Artificial Intelligence Nanodegree¶
Computer Vision Capstone¶
Project: Facial Keypoint Detection¶
Welcome to the final Computer Vision project in the Artificial Intelligence Nanodegree program!
In this project, you’ll combine your knowledge of computer vision techniques and deep learning to build and end-to-end facial keypoint recognition system! Facial keypoints include points around the eyes, nose, and mouth on any face and are used in many applications, from facial tracking to emotion recognition.
There are three main parts to this project:
Part 1 : Investigating OpenCV, pre-processing, and face detection
Part 2 : Training a Convolutional Neural Network (CNN) to detect facial keypoints
Part 3 : Putting parts 1 and 2 together to identify facial keypoints on any image!
*Here's what you need to know to complete the project:
In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested.
a. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation.
a. Each section where you will answer a question is preceded by a 'Question X' header.
b. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.
The rubric contains optional suggestions for enhancing the project beyond the minimum requirements. If you decide to pursue the "(Optional)" sections, you should include the code in this IPython notebook.
Your project submission will be evaluated based on your answers to each of the questions and the code implementations you provide.
Steps to Complete the Project¶
Each part of the notebook is further broken down into separate steps. Feel free to use the links below to navigate the notebook.
In this project you will get to explore a few of the many computer vision algorithms built into the OpenCV library. This expansive computer vision library is now almost 20 years old and still growing!
The project itself is broken down into three large parts, then even further into separate steps. Make sure to read through each step, and complete any sections that begin with '(IMPLEMENTATION)' in the header; these implementation sections may contain multiple TODOs that will be marked in code. For convenience, we provide links to each of these steps below.
Part 1 : Investigating OpenCV, pre-processing, and face detection
- Step 0: Detect Faces Using a Haar Cascade Classifier
- Step 1: Add Eye Detection
- Step 2: De-noise an Image for Better Face Detection
- Step 3: Blur an Image and Perform Edge Detection
- Step 4: Automatically Hide the Identity of an Individual
Part 2 : Training a Convolutional Neural Network (CNN) to detect facial keypoints
- Step 5: Create a CNN to Recognize Facial Keypoints
- Step 6: Compile and Train the Model
- Step 7: Visualize the Loss and Answer Questions
Part 3 : Putting parts 1 and 2 together to identify facial keypoints on any image!
- Step 8: Build a Robust Facial Keypoints Detector (Complete the CV Pipeline)
Step 0: Detect Faces Using a Haar Cascade Classifier¶
Have you ever wondered how Facebook automatically tags images with your friends' faces? Or how high-end cameras automatically find and focus on a certain person's face? Applications like these depend heavily on the machine learning task known as face detection - which is the task of automatically finding faces in images containing people.
At its root face detection is a classification problem - that is a problem of distinguishing between distinct classes of things. With face detection these distinct classes are 1) images of human faces and 2) everything else.
We use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images. OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the detector_architectures directory.
Import Resources¶
In the next python cell, we load in the required libraries for this section of the project.
In [2]:
# Import required libraries for this section
%matplotlib inline
import numpy as np
import matplotlib.pyplot as plt
import math
import cv2 # OpenCV library for computer vision
from PIL import Image
import time
Next, we load in and display a test image for performing face detection.
Note: by default OpenCV assumes the ordering of our image's color channels are Blue, then Green, then Red. This is slightly out of order with most image types we'll use in these experiments, whose color channels are ordered Red, then Green, then Blue. In order to switch the Blue and Red channels of our test image around we will use OpenCV's cvtColor function, which you can read more about by checking out some of its documentation located here. This is a general utility function that can do other transformations too like converting a color image to grayscale, and transforming a standard color image to HSV color space.
In [3]:
# Load in color image for face detection
image = cv2.imread('images/test_image_1.jpg')
# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
# Plot our image using subplots to specify a size and title
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Original Image')
ax1.imshow(image)
Out[3]:
<matplotlib.image.AxesImage at 0x7fe66b404160>
There are a lot of people - and faces - in this picture. 13 faces to be exact! In the next code cell, we demonstrate how to use a Haar Cascade classifier to detect all the faces in this test image.
This face detector uses information about patterns of intensity in an image to reliably detect faces under varying light conditions. So, to use this face detector, we'll first convert the image from color to grayscale.
Then, we load in the fully trained architecture of the face detector -- found in the file haarcascade_frontalface_default.xml - and use it on our image to find faces!
To learn more about the parameters of the detector see this post.
In [3]:
# Convert the RGB image to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)
# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')
# Detect the faces in image
faces = face_cascade.detectMultiScale(gray, 4, 6)
# Print the number of faces detected in the image
print('Number of faces detected:', len(faces))
# Make a copy of the orginal image to draw face detections on
image_with_detections = np.copy(image)
# Get the bounding box for each detected face
for (x,y,w,h) in faces:
# Add a red bounding box to the detections image
cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
# Display the image with the detections
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Image with Face Detections')
ax1.imshow(image_with_detections)
Number of faces detected: 13
Out[3]:
<matplotlib.image.AxesImage at 0x7efc8a8b4320>
In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.
Step 1: Add Eye Detections¶
There are other pre-trained detectors available that use a Haar Cascade Classifier - including full human body detectors, license plate detectors, and more. A full list of the pre-trained architectures can be found here.
To test your eye detector, we'll first read in a new test image with just a single face.
In [4]:
# Load in color image for face detection
image = cv2.imread('images/james.jpg')
# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
# Plot the RGB image
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Original Image')
ax1.imshow(image)
Out[4]:
<matplotlib.image.AxesImage at 0x7fe66b359358>
Notice that even though the image is a black and white image, we have read it in as a color image and so it will still need to be converted to grayscale in order to perform the most accurate face detection.
So, the next steps will be to convert this image to grayscale, then load OpenCV's face detector and run it with parameters that detect this face accurately.
In [5]:
# Convert the RGB image to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)
# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')
# Detect the faces in image
faces = face_cascade.detectMultiScale(gray, 1.25, 6)
# Print the number of faces detected in the image
print('Number of faces detected:', len(faces))
# Make a copy of the orginal image to draw face detections on
image_with_detections = np.copy(image)
# Get the bounding box for each detected face
for (x,y,w,h) in faces:
# Add a red bounding box to the detections image
cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
# Display the image with the detections
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Image with Face Detection')
ax1.imshow(image_with_detections)
Number of faces detected: 1
Out[5]:
<matplotlib.image.AxesImage at 0x7fe66b30c828>
(IMPLEMENTATION) Add an eye detector to the current face detection setup.¶
A Haar-cascade eye detector can be included in the same way that the face detector was and, in this first task, it will be your job to do just this.
To set up an eye detector, use the stored parameters of the eye cascade detector, called haarcascade_eye.xml, located in the detector_architectures subdirectory. In the next code cell, create your eye detector and store its detections.
A few notes before you get started:
First, make sure to give your loaded eye detector the variable name
eye_cascade
and give the list of eye regions you detect the variable name
eyes
Second, since we've already run the face detector over this image, you should only search for eyes within the rectangular face regions detected in faces. This will minimize false detections.
Lastly, once you've run your eye detector over the facial detection region, you should display the RGB image with both the face detection boxes (in red) and your eye detections (in green) to verify that everything works as expected.
In [6]:
# Make a copy of the original image to plot rectangle detections
image_with_detections = np.copy(image)
# Loop over the detections and draw their corresponding face detection boxes
for (x,y,w,h) in faces:
cv2.rectangle(image_with_detections, (x,y), (x+w,y+h),(255,0,0), 3)
# Do not change the code above this comment!
## Add eye detection, using haarcascade_eye.xml, to the current face detector algorithm
## Loop over the eye detections and draw their corresponding boxes in green on image_with_detections
eye_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_eye.xml')
for (x,y,w,h) in faces:
face_gray = gray[y:y+h, x:x+w]
eyes = eye_cascade.detectMultiScale(face_gray)
for (ex,ey,ew,eh) in eyes:
cv2.rectangle(image_with_detections, (x+ex,y+ey), (x+ex+ew,y+ey+eh), (0,255,0), 3)
# Plot the image with both faces and eyes detected
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Image with Face and Eye Detection')
ax1.imshow(image_with_detections)
Out[6]:
<matplotlib.image.AxesImage at 0x7fe66b338a20>
(Optional) Add face and eye detection to your laptop camera¶
It's time to kick it up a notch, and add face and eye detection to your laptop's camera! Afterwards, you'll be able to show off your creation like in the gif shown below - made with a completed version of the code!
Notice that not all of the detections here are perfect - and your result need not be perfect either. You should spend a small amount of time tuning the parameters of your detectors to get reasonable results, but don't hold out for perfection. If we wanted perfection we'd need to spend a ton of time tuning the parameters of each detector, cleaning up the input image frames, etc. You can think of this as more of a rapid prototype.
The next cell contains code for a wrapper function called laptop_camera_face_eye_detector that, when called, will activate your laptop's camera. You will place the relevant face and eye detection code in this wrapper function to implement face/eye detection and mark those detections on each image frame that your camera captures.
Before adding anything to the function, you can run it to get an idea of how it works - a small window should pop up showing you the live feed from your camera; you can press any key to close this window.
Note: Mac users may find that activating this function kills the kernel of their notebook every once in a while. If this happens to you, just restart your notebook's kernel, activate cell(s) containing any crucial import statements, and you'll be good to go!
In [8]:
### Add face and eye detection to this laptop camera function
# Make sure to draw out all faces/eyes found in each frame on the shown video feed
import cv2
import time
# wrapper function for face/eye detection with your laptop camera
def laptop_camera_go():
# Create instance of video capturer
cv2.namedWindow("face detection activated")
vc = cv2.VideoCapture(0)
# Try to get the first frame
if vc.isOpened():
rval, frame = vc.read()
else:
rval = False
# Keep the video stream open
while rval:
# Plot the image from camera with all the face and eye detections marked
cv2.imshow("face detection activated", frame)
# Exit functionality - press any key to exit laptop video
key = cv2.waitKey(20)
if key > 0: # Exit by pressing any key
# Destroy windows
cv2.destroyAllWindows()
# Make sure window closes on OSx
for i in range (1,5):
cv2.waitKey(1)
return
# Read next frame
time.sleep(0.05) # control framerate for computation - default 20 frames per sec
rval, frame = vc.read()
In [ ]:
# Call the laptop camera face/eye detector function above
laptop_camera_go()
Image quality is an important aspect of any computer vision task. Typically, when creating a set of images to train a deep learning network, significant care is taken to ensure that training images are free of visual noise or artifacts that hinder object detection. While computer vision algorithms - like a face detector - are typically trained on 'nice' data such as this, new test data doesn't always look so nice!
When applying a trained computer vision algorithm to a new piece of test data one often cleans it up first before feeding it in. This sort of cleaning - referred to as pre-processing - can include a number of cleaning phases like blurring, de-noising, color transformations, etc., and many of these tasks can be accomplished using OpenCV.
In this short subsection we explore OpenCV's noise-removal functionality to see how we can clean up a noisy image, which we then feed into our trained face detector.
Create a noisy image to work with¶
In the next cell, we create an artificial noisy version of the previous multi-face image. This is a little exaggerated - we don't typically get images that are this noisy - but image noise, or 'grainy-ness' in a digitial image - is a fairly common phenomenon.
In [7]:
# Load in the multi-face test image again
image = cv2.imread('images/test_image_1.jpg')
# Convert the image copy to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
# Make an array copy of this image
image_with_noise = np.asarray(image)
# Create noise - here we add noise sampled randomly from a Gaussian distribution: a common model for noise
noise_level = 40
noise = np.random.randn(image.shape[0],image.shape[1],image.shape[2])*noise_level
# Add this noise to the array image copy
image_with_noise = image_with_noise + noise
# Convert back to uint8 format
image_with_noise = np.asarray([np.uint8(np.clip(i,0,255)) for i in image_with_noise])
# Plot our noisy image!
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Noisy Image')
ax1.imshow(image_with_noise)
Out[7]:
<matplotlib.image.AxesImage at 0x7fe66b2ddcf8>
In the context of face detection, the problem with an image like this is that - due to noise - we may miss some faces or get false detections.
In the next cell we apply the same trained OpenCV detector with the same settings as before, to see what sort of detections we get.
In [10]:
# Convert the RGB image to grayscale
gray_noise = cv2.cvtColor(image_with_noise, cv2.COLOR_RGB2GRAY)
# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')
# Detect the faces in image
faces = face_cascade.detectMultiScale(gray_noise, 4, 6)
# Print the number of faces detected in the image
print('Number of faces detected:', len(faces))
# Make a copy of the orginal image to draw face detections on
image_with_detections = np.copy(image_with_noise)
# Get the bounding box for each detected face
for (x,y,w,h) in faces:
# Add a red bounding box to the detections image
cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
# Display the image with the detections
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Noisy Image with Face Detections')
ax1.imshow(image_with_detections)
Number of faces detected: 13
Out[10]:
<matplotlib.image.AxesImage at 0x7fe66b2b0ef0>
With this added noise we now miss one of the faces!
(IMPLEMENTATION) De-noise this image for better face detection¶
Time to get your hands dirty: using OpenCV's built in color image de-noising functionality called fastNlMeansDenoisingColored - de-noise this image enough so that all the faces in the image are properly detected. Once you have cleaned the image in the next cell, use the cell that follows to run our trained face detector over the cleaned image to check out its detections.
You can find its official documentation here and a useful example here.
Note: you can keep all parameters except photo_render fixed as shown in the second link above. Play around with the value of this parameter - see how it affects the resulting cleaned image.
In [16]:
## Use OpenCV's built in color image de-noising function to clean up our noisy image!
denoised_image = cv2.fastNlMeansDenoisingColored(image_with_noise,None,h = 12, hColor = 14,templateWindowSize = 7,searchWindowSize = 21)
In [17]:
## Run the face detector on the de-noised image to improve your detections and display the result
# Convert the RGB image to grayscale
gray_denoised = cv2.cvtColor(denoised_image, cv2.COLOR_RGB2GRAY)
# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')
# Detect the faces in image
faces = face_cascade.detectMultiScale(gray_denoised, 4, 6)
image_with_detections = np.copy(denoised_image)
# Get the bounding box for each detected face
for (x,y,w,h) in faces:
# Add a red bounding box to the detections image
cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
# Display the image with the detections
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Noisy Image with Face Detections')
ax1.imshow(image_with_detections)
Out[17]:
<matplotlib.image.AxesImage at 0x7fe62582b630>
Now that we have developed a simple pipeline for detecting faces using OpenCV - let's start playing around with a few fun things we can do with all those detected faces!
Importance of Blur in Edge Detection¶
Edge detection is a concept that pops up almost everywhere in computer vision applications, as edge-based features (as well as features built on top of edges) are often some of the best features for e.g., object detection and recognition problems.
Edge detection is a dimension reduction technique - by keeping only the edges of an image we get to throw away a lot of non-discriminating information. And typically the most useful kind of edge-detection is one that preserves only the important, global structures (ignoring local structures that aren't very discriminative). So removing local structures / retaining global structures is a crucial pre-processing step to performing edge detection in an image, and blurring can do just that.
Below is an animated gif showing the result of an edge-detected cat taken from Wikipedia, where the image is gradually blurred more and more prior to edge detection. When the animation begins you can't quite make out what it's a picture of, but as the animation evolves and local structures are removed via blurring the cat becomes visible in the edge-detected image.
Edge detection is a convolution performed on the image itself, and you can read about Canny edge detection on this OpenCV documentation page.
Canny edge detection¶
In the cell below we load in a test image, then apply Canny edge detection on it. The original image is shown on the left panel of the figure, while the edge-detected version of the image is shown on the right. Notice how the result looks very busy - there are too many little details preserved in the image before it is sent to the edge detector. When applied in computer vision applications, edge detection should preserve global structure; doing away with local structures that don't help describe what objects are in the image.
In [18]:
# Load in the image
image = cv2.imread('images/fawzia.jpg')
# Convert to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
# Convert to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)
# Perform Canny edge detection
edges = cv2.Canny(gray,100,200)
# Dilate the image to amplify edges
edges = cv2.dilate(edges, None)
# Plot the RGB and edge-detected image
fig = plt.figure(figsize = (15,15))
ax1 = fig.add_subplot(121)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Original Image')
ax1.imshow(image)
ax2 = fig.add_subplot(122)
ax2.set_xticks([])
ax2.set_yticks([])
ax2.set_title('Canny Edges')
ax2.imshow(edges, cmap='gray')
Out[18]:
<matplotlib.image.AxesImage at 0x7fe6257fd4e0>
Without first blurring the image, and removing small, local structures, a lot of irrelevant edge content gets picked up and amplified by the detector (as shown in the right panel above).
(IMPLEMENTATION) Blur the image then perform edge detection¶
In the next cell, you will repeat this experiment - blurring the image first to remove these local structures, so that only the important boudnary details remain in the edge-detected image.
Blur the image by using OpenCV's filter2d functionality - which is discussed in this documentation page - and use an averaging kernel of width equal to 4.
In [19]:
### Blur the test imageusing OpenCV's filter2d functionality,
# Use an averaging kernel, and a kernel width equal to 4
kernel = np.ones((4,4), np.float32) / 16
blured_image = cv2.filter2D(image, -1, kernel)
## Then perform Canny edge detection and display the output
# Convert to grayscale
gray = cv2.cvtColor(blured_image, cv2.COLOR_RGB2GRAY)
# Perform Canny edge detection
edges = cv2.Canny(gray,100,200)
# Dilate the image to amplify edges
edges = cv2.dilate(edges, None)
# Plot the RGB and edge-detected image
fig = plt.figure(figsize = (15,15))
ax1 = fig.add_subplot(121)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Original Image')
ax1.imshow(image)
ax2 = fig.add_subplot(122)
ax2.set_xticks([])
ax2.set_yticks([])
ax2.set_title('Canny Edges')
ax2.imshow(edges, cmap='gray')
Out[19]:
<matplotlib.image.AxesImage at 0x7fe6257d2080>
Step 4: Automatically Hide the Identity of an Individual¶
If you film something like a documentary or reality TV, you must get permission from every individual shown on film before you can show their face, otherwise you need to blur it out - by blurring the face a lot (so much so that even the global structures are obscured)! This is also true for projects like Google's StreetView maps - an enormous collection of mapping images taken from a fleet of Google vehicles. Because it would be impossible for Google to get the permission of every single person accidentally captured in one of these images they blur out everyone's faces, the detected images must automatically blur the identity of detected people. Here's a few examples of folks caught in the camera of a Google street view vehicle.
Read in an image to perform identity detection¶
Let's try this out for ourselves. Use the face detection pipeline built above and what you know about using the filter2D to blur and image, and use these in tandem to hide the identity of the person in the following image - loaded in and printed in the next cell.
In [20]:
# Load in the image
image = cv2.imread('images/gus.jpg')
# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
# Display the image
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Original Image')
ax1.imshow(image)
Out[20]:
<matplotlib.image.AxesImage at 0x7fe625776908>
(IMPLEMENTATION) Use blurring to hide the identity of an individual in an image¶
The idea here is to 1) automatically detect the face in this image, and then 2) blur it out! Make sure to adjust the parameters of the averaging blur filter to completely obscure this person's identity.
In [21]:
## Implement face detection
gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)
# Detect the faces in image
faces = face_cascade.detectMultiScale(gray, 1.07, 20)
## Blur the bounding box around each detected face using an averaging filter and display the result
image_blured = np.copy(image)
# Blur each face
for (x,y,w,h) in faces:
face = image_blured[y:y+h, x:x+w]
image_blured[y:y+h, x:x+w] = cv2.blur(face,(100,100))
# Display the image
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Image with blured face')
ax1.imshow(image_blured)
Out[21]:
<matplotlib.image.AxesImage at 0x7fe62571ff28>
(Optional) Build identity protection into your laptop camera¶
In this optional task you can add identity protection to your laptop camera, using the previously completed code where you added face detection to your laptop camera - and the task above. You should be able to get reasonable results with little parameter tuning - like the one shown in the gif below.
As with the previous video task, to make this perfect would require significant effort - so don't strive for perfection here, strive for reasonable quality.
The next cell contains code a wrapper function called laptop_camera_identity_hider that - when called - will activate your laptop's camera. You need to place the relevant face detection and blurring code developed above in this function in order to blur faces entering your laptop camera's field of view.
Before adding anything to the function you can call it to get a hang of how it works - a small window will pop up showing you the live feed from your camera, you can press any key to close this window.
Note: Mac users may find that activating this function kills the kernel of their notebook every once in a while. If this happens to you, just restart your notebook's kernel, activate cell(s) containing any crucial import statements, and you'll be good to go!
In [ ]:
### Insert face detection and blurring code into the wrapper below to create an identity protector on your laptop!
import cv2
import time
def laptop_camera_go():
# Create instance of video capturer
cv2.namedWindow("face detection activated")
vc = cv2.VideoCapture(0)
# Try to get the first frame
if vc.isOpened():
rval, frame = vc.read()
else:
rval = False
# Keep video stream open
while rval:
# Plot image from camera with detections marked
cv2.imshow("face detection activated", frame)
# Exit functionality - press any key to exit laptop video
key = cv2.waitKey(20)
if key > 0: # Exit by pressing any key
# Destroy windows
cv2.destroyAllWindows()
for i in range (1,5):
cv2.waitKey(1)
return
# Read next frame
time.sleep(0.05) # control framerate for computation - default 20 frames per sec
rval, frame = vc.read()
In [ ]:
# Run laptop identity hider
laptop_camera_go()
Step 5: Create a CNN to Recognize Facial Keypoints¶
OpenCV is often used in practice with other machine learning and deep learning libraries to produce interesting results. In this stage of the project you will create your own end-to-end pipeline - employing convolutional networks in keras along with OpenCV - to apply a "selfie" filter to streaming video and images.
You will start by creating and then training a convolutional network that can detect facial keypoints in a small dataset of cropped images of human faces. We then guide you towards OpenCV to expanding your detection algorithm to more general images. What are facial keypoints? Let's take a look at some examples.
Facial keypoints (also called facial landmarks) are the small blue-green dots shown on each of the faces in the image above - there are 15 keypoints marked in each image. They mark important areas of the face - the eyes, corners of the mouth, the nose, etc. Facial keypoints can be used in a variety of machine learning applications from face and emotion recognition to commercial applications like the image filters popularized by Snapchat.
Below we illustrate a filter that, using the results of this section, automatically places sunglasses on people in images (using the facial keypoints to place the glasses correctly on each face). Here, the facial keypoints have been colored lime green for visualization purposes.
Make a facial keypoint detector¶
But first things first: how can we make a facial keypoint detector? Well, at a high level, notice that facial keypoint detection is a regression problem. A single face corresponds to a set of 15 facial keypoints (a set of 15 corresponding $(x, y)$ coordinates, i.e., an output point). Because our input data are images, we can employ a convolutional neural network to recognize patterns in our images and learn how to identify these keypoint given sets of labeled data.
In order to train a regressor, we need a training set - a set of facial image / facial keypoint pairs to train on. For this we will be using this dataset from Kaggle. We've already downloaded this data and placed it in the data directory. Make sure that you have both the training and test data files. The training dataset contains several thousand $96 \times 96$ grayscale images of cropped human faces, along with each face's 15 corresponding facial keypoints (also called landmarks) that have been placed by hand, and recorded in $(x, y)$ coordinates. This wonderful resource also has a substantial testing set, which we will use in tinkering with our convolutional network.
To load in this data, run the Python cell below - notice we will load in both the training and testing sets.
The load_data function is in the included utils.py file.
In [1]:
from utils import *
# Load training set
X_train, y_train = load_data()
print("X_train.shape == {}".format(X_train.shape))
print("y_train.shape == {}; y_train.min == {:.3f}; y_train.max == {:.3f}".format(
y_train.shape, y_train.min(), y_train.max()))
# Load testing set
X_test, _ = load_data(test=True)
print("X_test.shape == {}".format(X_test.shape))
Using TensorFlow backend.
X_train.shape == (2140, 96, 96, 1) y_train.shape == (2140, 30); y_train.min == -0.920; y_train.max == 0.996 X_test.shape == (1783, 96, 96, 1)
The load_data function in utils.py originates from this excellent blog post, which you are strongly encouraged to read. Please take the time now to review this function. Note how the output values - that is, the coordinates of each set of facial landmarks - have been normalized to take on values in the range $[-1, 1]$, while the pixel values of each input point (a facial image) have been normalized to the range $[0,1]$.
Note: the original Kaggle dataset contains some images with several missing keypoints. For simplicity, the load_data function removes those images with missing labels from the dataset. As an optional extension, you are welcome to amend the load_data function to include the incomplete data points.
Visualize the Training Data¶
Execute the code cell below to visualize a subset of the training data.
In [2]:
import matplotlib.pyplot as plt
%matplotlib inline
fig = plt.figure(figsize=(20,20))
fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)
for i in range(9):
ax = fig.add_subplot(3, 3, i + 1, xticks=[], yticks=[])
plot_data(X_train[i], y_train[i], ax)
For each training image, there are two landmarks per eyebrow (four total), three per eye (six total), four for the mouth, and one for the tip of the nose.
Review the plot_data function in utils.py to understand how the 30-dimensional training labels in y_train are mapped to facial locations, as this function will prove useful for your pipeline.
(IMPLEMENTATION) Specify the CNN Architecture¶
In this section, you will specify a neural network for predicting the locations of facial keypoints. Use the code cell below to specify the architecture of your neural network. We have imported some layers that you may find useful for this task, but if you need to use more Keras layers, feel free to import them in the cell.
Your network should accept a $96 \times 96$ grayscale image as input, and it should output a vector with 30 entries, corresponding to the predicted (horizontal and vertical) locations of 15 facial keypoints. If you are not sure where to start, you can find some useful starting architectures in this blog, but you are not permitted to copy any of the architectures that you find online.
In [59]:
# Import deep learning resources from Keras
from keras.models import Sequential
from keras.layers import Conv2D, MaxPooling2D, Dropout
from keras.layers import Flatten, Dense
## Specify a CNN architecture
# Your model should accept 96x96 pixel graysale images in
# It should have a fully-connected output layer with 30 values (2 for each facial keypoint)
model = Sequential()
model.add(Conv2D(32, (4, 4), activation='relu', input_shape=(96, 96, 1)))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.3))
model.add(Conv2D(128, (2, 2), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.4))
model.add(Flatten())
model.add(Dense(1024, activation='tanh'))
model.add(Dropout(0.5))
model.add(Dense(512, activation='tanh'))
model.add(Dropout(0.5))
model.add(Dense(30, activation='tanh'))
# Summarize the model
model.summary()
_________________________________________________________________ Layer (type) Output Shape Param # ================================================================= conv2d_34 (Conv2D) (None, 93, 93, 32) 544 _________________________________________________________________ max_pooling2d_34 (MaxPooling (None, 46, 46, 32) 0 _________________________________________________________________ conv2d_35 (Conv2D) (None, 44, 44, 64) 18496 _________________________________________________________________ max_pooling2d_35 (MaxPooling (None, 22, 22, 64) 0 _________________________________________________________________ dropout_45 (Dropout) (None, 22, 22, 64) 0 _________________________________________________________________ conv2d_36 (Conv2D) (None, 21, 21, 128) 32896 _________________________________________________________________ max_pooling2d_36 (MaxPooling (None, 10, 10, 128) 0 _________________________________________________________________ dropout_46 (Dropout) (None, 10, 10, 128) 0 _________________________________________________________________ flatten_12 (Flatten) (None, 12800) 0 _________________________________________________________________ dense_34 (Dense) (None, 1024) 13108224 _________________________________________________________________ dropout_47 (Dropout) (None, 1024) 0 _________________________________________________________________ dense_35 (Dense) (None, 512) 524800 _________________________________________________________________ dropout_48 (Dropout) (None, 512) 0 _________________________________________________________________ dense_36 (Dense) (None, 30) 15390 ================================================================= Total params: 13,700,350 Trainable params: 13,700,350 Non-trainable params: 0 _________________________________________________________________
Step 6: Compile and Train the Model¶
After specifying your architecture, you'll need to compile and train the model to detect facial keypoints'
(IMPLEMENTATION) Compile and Train the Model¶
Use the compile method to configure the learning process. Experiment with your choice of optimizer; you may have some ideas about which will work best (SGD vs. RMSprop, etc), but take the time to empirically verify your theories.
Use the fit method to train the model. Break off a validation set by setting validation_split=0.2. Save the returned History object in the history variable.
Experiment with your model to minimize the validation loss (measured as mean squared error). A very good model will achieve about 0.0015 loss (though it's possible to do even better). When you have finished training, save your model as an HDF5 file with file path my_model.h5.
In [60]:
from keras.optimizers import SGD, RMSprop, Adagrad, Adadelta, Adam, Adamax, Nadam
## Compile the model
# sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
# adam = Adam(lr=0.01, beta_1=0.9, beta_2=0.999, decay=1e-6)
# nadam = Nadam(lr=0.002, beta_1=0.9, beta_2=0.999, schedule_decay=0.004)
model.compile(optimizer='rmsprop', loss='mean_squared_error', metrics=['accuracy'])
## Train the model
history = model.fit(X_train, y_train, batch_size=128, epochs=3000, validation_split=0.2)
## TODO: Save the model as model.h5
model.save('my_model.h5')
Train on 1712 samples, validate on 428 samples Epoch 1/3000 1712/1712 [==============================] - 2s - loss: 0.1533 - acc: 0.3359 - val_loss: 0.0094 - val_acc: 0.6963 Epoch 2/3000 1712/1712 [==============================] - 1s - loss: 0.0181 - acc: 0.5047 - val_loss: 0.0054 - val_acc: 0.6963 Epoch 3/3000 1712/1712 [==============================] - 1s - loss: 0.0121 - acc: 0.5607 - val_loss: 0.0046 - val_acc: 0.6963 Epoch 4/3000 1712/1712 [==============================] - 1s - loss: 0.0095 - acc: 0.6063 - val_loss: 0.0049 - val_acc: 0.6963 Epoch 5/3000 1712/1712 [==============================] - 1s - loss: 0.0087 - acc: 0.6373 - val_loss: 0.0096 - val_acc: 0.6963 Epoch 6/3000 1712/1712 [==============================] - 1s - loss: 0.0128 - acc: 0.6624 - val_loss: 0.0048 - val_acc: 0.6963 Epoch 7/3000 1712/1712 [==============================] - 1s - loss: 0.0084 - acc: 0.6706 - val_loss: 0.0055 - val_acc: 0.6963 Epoch 8/3000 1712/1712 [==============================] - 1s - loss: 0.0078 - acc: 0.6840 - val_loss: 0.0048 - val_acc: 0.6963 Epoch 9/3000 1712/1712 [==============================] - 1s - loss: 0.0077 - acc: 0.6787 - val_loss: 0.0054 - val_acc: 0.6963 Epoch 10/3000 1712/1712 [==============================] - 1s - loss: 0.0076 - acc: 0.6916 - val_loss: 0.0068 - val_acc: 0.6963 Epoch 11/3000 1712/1712 [==============================] - 1s - loss: 0.0077 - acc: 0.6951 - val_loss: 0.0043 - val_acc: 0.6963 Epoch 12/3000 1712/1712 [==============================] - 1s - loss: 0.0064 - acc: 0.6928 - val_loss: 0.0045 - val_acc: 0.6963 Epoch 13/3000 1712/1712 [==============================] - 1s - loss: 0.0076 - acc: 0.6992 - val_loss: 0.0043 - val_acc: 0.6963 Epoch 14/3000 1712/1712 [==============================] - 1s - loss: 0.0056 - acc: 0.6992 - val_loss: 0.0046 - val_acc: 0.6963 Epoch 15/3000 1712/1712 [==============================] - 1s - loss: 0.0062 - acc: 0.7039 - val_loss: 0.0040 - val_acc: 0.7009 Epoch 16/3000 1712/1712 [==============================] - 1s - loss: 0.0054 - acc: 0.6968 - val_loss: 0.0047 - val_acc: 0.6963 Epoch 17/3000 1712/1712 [==============================] - 1s - loss: 0.0244 - acc: 0.6636 - val_loss: 0.0052 - val_acc: 0.6963 Epoch 18/3000 1712/1712 [==============================] - 1s - loss: 0.0054 - acc: 0.7050 - val_loss: 0.0042 - val_acc: 0.6963 Epoch 19/3000 1712/1712 [==============================] - 1s - loss: 0.0051 - acc: 0.6939 - val_loss: 0.0058 - val_acc: 0.6963 Epoch 20/3000 1712/1712 [==============================] - 1s - loss: 0.0054 - acc: 0.7004 - val_loss: 0.0059 - val_acc: 0.6963 Epoch 21/3000 1712/1712 [==============================] - 1s - loss: 0.0051 - acc: 0.6986 - val_loss: 0.0040 - val_acc: 0.6963 Epoch 22/3000 1712/1712 [==============================] - 1s - loss: 0.0048 - acc: 0.7009 - val_loss: 0.0035 - val_acc: 0.6986 Epoch 23/3000 1712/1712 [==============================] - 1s - loss: 0.0046 - acc: 0.6974 - val_loss: 0.0040 - val_acc: 0.7009 Epoch 24/3000 1712/1712 [==============================] - 1s - loss: 0.0045 - acc: 0.7033 - val_loss: 0.0041 - val_acc: 0.6963 Epoch 25/3000 1712/1712 [==============================] - 1s - loss: 0.0044 - acc: 0.6957 - val_loss: 0.0031 - val_acc: 0.7009 Epoch 26/3000 1712/1712 [==============================] - 1s - loss: 0.0045 - acc: 0.7033 - val_loss: 0.0032 - val_acc: 0.6986 Epoch 27/3000 1712/1712 [==============================] - 1s - loss: 0.0042 - acc: 0.7091 - val_loss: 0.0037 - val_acc: 0.6986 Epoch 28/3000 1712/1712 [==============================] - 1s - loss: 0.0038 - acc: 0.7056 - val_loss: 0.0033 - val_acc: 0.6963 Epoch 29/3000 1712/1712 [==============================] - 1s - loss: 0.0041 - acc: 0.7056 - val_loss: 0.0027 - val_acc: 0.7033 Epoch 30/3000 1712/1712 [==============================] - 1s - loss: 0.0038 - acc: 0.7039 - val_loss: 0.0033 - val_acc: 0.7103 Epoch 31/3000 1712/1712 [==============================] - 1s - loss: 0.0038 - acc: 0.7109 - val_loss: 0.0039 - val_acc: 0.7150 Epoch 32/3000 1712/1712 [==============================] - 1s - loss: 0.0038 - acc: 0.7068 - val_loss: 0.0031 - val_acc: 0.7126 Epoch 33/3000 1712/1712 [==============================] - 1s - loss: 0.0035 - acc: 0.7015 - val_loss: 0.0038 - val_acc: 0.6963 Epoch 34/3000 1712/1712 [==============================] - 1s - loss: 0.0035 - acc: 0.7050 - val_loss: 0.0025 - val_acc: 0.6986 Epoch 35/3000 1712/1712 [==============================] - 1s - loss: 0.0039 - acc: 0.7144 - val_loss: 0.0025 - val_acc: 0.7243 Epoch 36/3000 1712/1712 [==============================] - 1s - loss: 0.0032 - acc: 0.7144 - val_loss: 0.0024 - val_acc: 0.7033 Epoch 37/3000 1712/1712 [==============================] - 1s - loss: 0.0032 - acc: 0.7091 - val_loss: 0.0027 - val_acc: 0.7266 Epoch 38/3000 1712/1712 [==============================] - 1s - loss: 0.0035 - acc: 0.7085 - val_loss: 0.0042 - val_acc: 0.6916 Epoch 39/3000 1712/1712 [==============================] - 1s - loss: 0.0032 - acc: 0.7155 - val_loss: 0.0046 - val_acc: 0.6963 Epoch 40/3000 1712/1712 [==============================] - 1s - loss: 0.0034 - acc: 0.7214 - val_loss: 0.0028 - val_acc: 0.7103 Epoch 41/3000 1712/1712 [==============================] - 1s - loss: 0.0030 - acc: 0.7266 - val_loss: 0.0036 - val_acc: 0.7033 Epoch 42/3000 1712/1712 [==============================] - 1s - loss: 0.0033 - acc: 0.7255 - val_loss: 0.0031 - val_acc: 0.6963 Epoch 43/3000 1712/1712 [==============================] - 1s - loss: 0.0030 - acc: 0.7301 - val_loss: 0.0023 - val_acc: 0.7150 Epoch 44/3000 1712/1712 [==============================] - 1s - loss: 0.0029 - acc: 0.7348 - val_loss: 0.0026 - val_acc: 0.7079 Epoch 45/3000 1712/1712 [==============================] - 1s - loss: 0.0032 - acc: 0.7360 - val_loss: 0.0018 - val_acc: 0.7173 Epoch 46/3000 1712/1712 [==============================] - 1s - loss: 0.0033 - acc: 0.7401 - val_loss: 0.0022 - val_acc: 0.7313 Epoch 47/3000 1712/1712 [==============================] - 1s - loss: 0.0026 - acc: 0.7307 - val_loss: 0.0018 - val_acc: 0.7173 Epoch 48/3000 1712/1712 [==============================] - 1s - loss: 0.0030 - acc: 0.7336 - val_loss: 0.0021 - val_acc: 0.7220 Epoch 49/3000 1712/1712 [==============================] - 1s - loss: 0.0025 - acc: 0.7482 - val_loss: 0.0035 - val_acc: 0.6939 Epoch 50/3000 1712/1712 [==============================] - 1s - loss: 0.0031 - acc: 0.7412 - val_loss: 0.0025 - val_acc: 0.7009 Epoch 51/3000 1712/1712 [==============================] - 1s - loss: 0.0027 - acc: 0.7342 - val_loss: 0.0025 - val_acc: 0.6986 Epoch 52/3000 1712/1712 [==============================] - 1s - loss: 0.0027 - acc: 0.7342 - val_loss: 0.0027 - val_acc: 0.7056 Epoch 53/3000 1712/1712 [==============================] - 1s - loss: 0.0026 - acc: 0.7588 - val_loss: 0.0027 - val_acc: 0.7266 Epoch 54/3000 1712/1712 [==============================] - 1s - loss: 0.0030 - acc: 0.7383 - val_loss: 0.0025 - val_acc: 0.7079 Epoch 55/3000 1712/1712 [==============================] - 1s - loss: 0.0023 - acc: 0.7442 - val_loss: 0.0021 - val_acc: 0.7126 Epoch 56/3000 1712/1712 [==============================] - 1s - loss: 0.0029 - acc: 0.7418 - val_loss: 0.0020 - val_acc: 0.7290 Epoch 57/3000 1712/1712 [==============================] - 1s - loss: 0.0025 - acc: 0.7371 - val_loss: 0.0025 - val_acc: 0.7150 Epoch 58/3000 1712/1712 [==============================] - 1s - loss: 0.0025 - acc: 0.7518 - val_loss: 0.0032 - val_acc: 0.6986 Epoch 59/3000 1712/1712 [==============================] - 1s - loss: 0.0027 - acc: 0.7447 - val_loss: 0.0022 - val_acc: 0.7266 Epoch 60/3000 1712/1712 [==============================] - 1s - loss: 0.0024 - acc: 0.7570 - val_loss: 0.0025 - val_acc: 0.7103 Epoch 61/3000 1712/1712 [==============================] - 1s - loss: 0.0026 - acc: 0.7488 - val_loss: 0.0024 - val_acc: 0.7173 Epoch 62/3000 1712/1712 [==============================] - 1s - loss: 0.0023 - acc: 0.7564 - val_loss: 0.0031 - val_acc: 0.7150 Epoch 63/3000 1712/1712 [==============================] - 1s - loss: 0.0026 - acc: 0.7518 - val_loss: 0.0020 - val_acc: 0.7290 Epoch 64/3000 1712/1712 [==============================] - 1s - loss: 0.0024 - acc: 0.7547 - val_loss: 0.0024 - val_acc: 0.7079 Epoch 65/3000 1712/1712 [==============================] - 1s - loss: 0.0023 - acc: 0.7605 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 66/3000 1712/1712 [==============================] - 1s - loss: 0.0024 - acc: 0.7529 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 67/3000 1712/1712 [==============================] - 1s - loss: 0.0027 - acc: 0.7477 - val_loss: 0.0037 - val_acc: 0.7009 Epoch 68/3000 1712/1712 [==============================] - 1s - loss: 0.0025 - acc: 0.7389 - val_loss: 0.0020 - val_acc: 0.7103 Epoch 69/3000 1712/1712 [==============================] - 1s - loss: 0.0021 - acc: 0.7401 - val_loss: 0.0023 - val_acc: 0.7687 Epoch 70/3000 1712/1712 [==============================] - 1s - loss: 0.0023 - acc: 0.7436 - val_loss: 0.0017 - val_acc: 0.7570 Epoch 71/3000 1712/1712 [==============================] - 1s - loss: 0.0027 - acc: 0.7664 - val_loss: 0.0016 - val_acc: 0.7477 Epoch 72/3000 1712/1712 [==============================] - 1s - loss: 0.0021 - acc: 0.7477 - val_loss: 0.0022 - val_acc: 0.7360 Epoch 73/3000 1712/1712 [==============================] - 1s - loss: 0.0023 - acc: 0.7623 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 74/3000 1712/1712 [==============================] - 1s - loss: 0.0021 - acc: 0.7564 - val_loss: 0.0024 - val_acc: 0.7313 Epoch 75/3000 1712/1712 [==============================] - 1s - loss: 0.0024 - acc: 0.7687 - val_loss: 0.0022 - val_acc: 0.7266 Epoch 76/3000 1712/1712 [==============================] - 1s - loss: 0.0021 - acc: 0.7675 - val_loss: 0.0024 - val_acc: 0.6986 Epoch 77/3000 1712/1712 [==============================] - 1s - loss: 0.0023 - acc: 0.7553 - val_loss: 0.0025 - val_acc: 0.7196 Epoch 78/3000 1712/1712 [==============================] - 1s - loss: 0.0020 - acc: 0.7739 - val_loss: 0.0029 - val_acc: 0.7056 Epoch 79/3000 1712/1712 [==============================] - 1s - loss: 0.0023 - acc: 0.7512 - val_loss: 0.0018 - val_acc: 0.7360 Epoch 80/3000 1712/1712 [==============================] - 1s - loss: 0.0021 - acc: 0.7634 - val_loss: 0.0018 - val_acc: 0.7547 Epoch 81/3000 1712/1712 [==============================] - 1s - loss: 0.0021 - acc: 0.7605 - val_loss: 0.0025 - val_acc: 0.7477 Epoch 82/3000 1712/1712 [==============================] - 1s - loss: 0.0025 - acc: 0.7535 - val_loss: 0.0027 - val_acc: 0.6939 Epoch 83/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7652 - val_loss: 0.0026 - val_acc: 0.7150 Epoch 84/3000 1712/1712 [==============================] - 1s - loss: 0.0021 - acc: 0.7769 - val_loss: 0.0019 - val_acc: 0.7850 Epoch 85/3000 1712/1712 [==============================] - 1s - loss: 0.0021 - acc: 0.7757 - val_loss: 0.0019 - val_acc: 0.7173 Epoch 86/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7658 - val_loss: 0.0026 - val_acc: 0.7056 Epoch 87/3000 1712/1712 [==============================] - 1s - loss: 0.0021 - acc: 0.7564 - val_loss: 0.0017 - val_acc: 0.7734 Epoch 88/3000 1712/1712 [==============================] - 1s - loss: 0.0022 - acc: 0.7629 - val_loss: 0.0019 - val_acc: 0.7593 Epoch 89/3000 1712/1712 [==============================] - 1s - loss: 0.0020 - acc: 0.7728 - val_loss: 0.0024 - val_acc: 0.7196 Epoch 90/3000 1712/1712 [==============================] - 1s - loss: 0.0023 - acc: 0.7675 - val_loss: 0.0018 - val_acc: 0.7547 Epoch 91/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7763 - val_loss: 0.0024 - val_acc: 0.6986 Epoch 92/3000 1712/1712 [==============================] - 1s - loss: 0.0022 - acc: 0.7588 - val_loss: 0.0022 - val_acc: 0.7243 Epoch 93/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7664 - val_loss: 0.0023 - val_acc: 0.7103 Epoch 94/3000 1712/1712 [==============================] - 1s - loss: 0.0020 - acc: 0.7640 - val_loss: 0.0029 - val_acc: 0.7009 Epoch 95/3000 1712/1712 [==============================] - 1s - loss: 0.0023 - acc: 0.7669 - val_loss: 0.0015 - val_acc: 0.7243 Epoch 96/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7792 - val_loss: 0.0018 - val_acc: 0.7266 Epoch 97/3000 1712/1712 [==============================] - 1s - loss: 0.0021 - acc: 0.7658 - val_loss: 0.0020 - val_acc: 0.7453 Epoch 98/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7687 - val_loss: 0.0025 - val_acc: 0.7126 Epoch 99/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7810 - val_loss: 0.0019 - val_acc: 0.7220 Epoch 100/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7693 - val_loss: 0.0016 - val_acc: 0.7266 Epoch 101/3000 1712/1712 [==============================] - 1s - loss: 0.0022 - acc: 0.7646 - val_loss: 0.0018 - val_acc: 0.7196 Epoch 102/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7640 - val_loss: 0.0026 - val_acc: 0.7360 Epoch 103/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7693 - val_loss: 0.0020 - val_acc: 0.7336 Epoch 104/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7810 - val_loss: 0.0018 - val_acc: 0.7430 Epoch 105/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7757 - val_loss: 0.0019 - val_acc: 0.7523 Epoch 106/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7810 - val_loss: 0.0015 - val_acc: 0.7804 Epoch 107/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7769 - val_loss: 0.0031 - val_acc: 0.7033 Epoch 108/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7646 - val_loss: 0.0016 - val_acc: 0.7570 Epoch 109/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7792 - val_loss: 0.0020 - val_acc: 0.7360 Epoch 110/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7599 - val_loss: 0.0026 - val_acc: 0.7056 Epoch 111/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7775 - val_loss: 0.0023 - val_acc: 0.7009 Epoch 112/3000 1712/1712 [==============================] - 1s - loss: 0.0020 - acc: 0.7640 - val_loss: 0.0015 - val_acc: 0.7687 Epoch 113/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7675 - val_loss: 0.0019 - val_acc: 0.7220 Epoch 114/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7839 - val_loss: 0.0022 - val_acc: 0.7477 Epoch 115/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7699 - val_loss: 0.0018 - val_acc: 0.7547 Epoch 116/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7926 - val_loss: 0.0035 - val_acc: 0.7033 Epoch 117/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7856 - val_loss: 0.0016 - val_acc: 0.7453 Epoch 118/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7839 - val_loss: 0.0019 - val_acc: 0.7710 Epoch 119/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7582 - val_loss: 0.0020 - val_acc: 0.7103 Epoch 120/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7839 - val_loss: 0.0020 - val_acc: 0.7967 Epoch 121/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7775 - val_loss: 0.0020 - val_acc: 0.7220 Epoch 122/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7827 - val_loss: 0.0021 - val_acc: 0.7430 Epoch 123/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7880 - val_loss: 0.0020 - val_acc: 0.7757 Epoch 124/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7804 - val_loss: 0.0019 - val_acc: 0.7710 Epoch 125/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7903 - val_loss: 0.0027 - val_acc: 0.7079 Epoch 126/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7874 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 127/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7880 - val_loss: 0.0020 - val_acc: 0.7336 Epoch 128/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7833 - val_loss: 0.0017 - val_acc: 0.7664 Epoch 129/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7827 - val_loss: 0.0025 - val_acc: 0.7009 Epoch 130/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7886 - val_loss: 0.0017 - val_acc: 0.7243 Epoch 131/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7739 - val_loss: 0.0029 - val_acc: 0.7407 Epoch 132/3000 1712/1712 [==============================] - 1s - loss: 0.0019 - acc: 0.7921 - val_loss: 0.0017 - val_acc: 0.7453 Epoch 133/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7845 - val_loss: 0.0024 - val_acc: 0.6939 Epoch 134/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7833 - val_loss: 0.0014 - val_acc: 0.7617 Epoch 135/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7862 - val_loss: 0.0019 - val_acc: 0.7336 Epoch 136/3000 1712/1712 [==============================] - 1s - loss: 0.0018 - acc: 0.7792 - val_loss: 0.0022 - val_acc: 0.6986 Epoch 137/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7956 - val_loss: 0.0019 - val_acc: 0.7336 Epoch 138/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7786 - val_loss: 0.0015 - val_acc: 0.7523 Epoch 139/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7944 - val_loss: 0.0016 - val_acc: 0.7453 Epoch 140/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7845 - val_loss: 0.0027 - val_acc: 0.7009 Epoch 141/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7850 - val_loss: 0.0021 - val_acc: 0.7640 Epoch 142/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.8014 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 143/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7956 - val_loss: 0.0016 - val_acc: 0.8037 Epoch 144/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7944 - val_loss: 0.0018 - val_acc: 0.7453 Epoch 145/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7856 - val_loss: 0.0016 - val_acc: 0.7313 Epoch 146/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7833 - val_loss: 0.0017 - val_acc: 0.7804 Epoch 147/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7868 - val_loss: 0.0029 - val_acc: 0.7500 Epoch 148/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7996 - val_loss: 0.0015 - val_acc: 0.7360 Epoch 149/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7810 - val_loss: 0.0023 - val_acc: 0.7079 Epoch 150/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7891 - val_loss: 0.0014 - val_acc: 0.7313 Epoch 151/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7868 - val_loss: 0.0021 - val_acc: 0.7033 Epoch 152/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7915 - val_loss: 0.0017 - val_acc: 0.7196 Epoch 153/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.8020 - val_loss: 0.0019 - val_acc: 0.7336 Epoch 154/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7839 - val_loss: 0.0022 - val_acc: 0.7850 Epoch 155/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7815 - val_loss: 0.0021 - val_acc: 0.6986 Epoch 156/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7932 - val_loss: 0.0022 - val_acc: 0.7150 Epoch 157/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7897 - val_loss: 0.0019 - val_acc: 0.7220 Epoch 158/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7868 - val_loss: 0.0017 - val_acc: 0.7757 Epoch 159/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7909 - val_loss: 0.0027 - val_acc: 0.7126 Epoch 160/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7786 - val_loss: 0.0014 - val_acc: 0.7757 Epoch 161/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.8090 - val_loss: 0.0015 - val_acc: 0.7687 Epoch 162/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7973 - val_loss: 0.0018 - val_acc: 0.7897 Epoch 163/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7932 - val_loss: 0.0017 - val_acc: 0.7290 Epoch 164/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.8014 - val_loss: 0.0020 - val_acc: 0.7734 Epoch 165/3000 1712/1712 [==============================] - 1s - loss: 0.0017 - acc: 0.7956 - val_loss: 0.0021 - val_acc: 0.7827 Epoch 166/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7775 - val_loss: 0.0015 - val_acc: 0.8037 Epoch 167/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7891 - val_loss: 0.0019 - val_acc: 0.7523 Epoch 168/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7956 - val_loss: 0.0014 - val_acc: 0.7547 Epoch 169/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.8020 - val_loss: 0.0015 - val_acc: 0.7593 Epoch 170/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7921 - val_loss: 0.0015 - val_acc: 0.7897 Epoch 171/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7903 - val_loss: 0.0023 - val_acc: 0.7009 Epoch 172/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.8020 - val_loss: 0.0019 - val_acc: 0.7430 Epoch 173/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7926 - val_loss: 0.0021 - val_acc: 0.7196 Epoch 174/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7909 - val_loss: 0.0025 - val_acc: 0.7220 Epoch 175/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.8125 - val_loss: 0.0015 - val_acc: 0.7687 Epoch 176/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7874 - val_loss: 0.0014 - val_acc: 0.8154 Epoch 177/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7845 - val_loss: 0.0019 - val_acc: 0.7103 Epoch 178/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7915 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 179/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7991 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 180/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.8020 - val_loss: 0.0016 - val_acc: 0.8014 Epoch 181/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7950 - val_loss: 0.0019 - val_acc: 0.7150 Epoch 182/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7921 - val_loss: 0.0016 - val_acc: 0.7383 Epoch 183/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7991 - val_loss: 0.0021 - val_acc: 0.7056 Epoch 184/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7874 - val_loss: 0.0030 - val_acc: 0.6939 Epoch 185/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7786 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 186/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.8096 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 187/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8213 - val_loss: 0.0021 - val_acc: 0.7266 Epoch 188/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.8113 - val_loss: 0.0015 - val_acc: 0.7967 Epoch 189/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.8078 - val_loss: 0.0019 - val_acc: 0.7804 Epoch 190/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.8102 - val_loss: 0.0023 - val_acc: 0.7103 Epoch 191/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7967 - val_loss: 0.0016 - val_acc: 0.7734 Epoch 192/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7944 - val_loss: 0.0022 - val_acc: 0.7266 Epoch 193/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.8026 - val_loss: 0.0018 - val_acc: 0.7243 Epoch 194/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.7950 - val_loss: 0.0018 - val_acc: 0.7687 Epoch 195/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7915 - val_loss: 0.0019 - val_acc: 0.7360 Epoch 196/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8008 - val_loss: 0.0016 - val_acc: 0.7500 Epoch 197/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7985 - val_loss: 0.0017 - val_acc: 0.7967 Epoch 198/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.7979 - val_loss: 0.0014 - val_acc: 0.7921 Epoch 199/3000 1712/1712 [==============================] - 1s - loss: 0.0015 - acc: 0.7996 - val_loss: 0.0015 - val_acc: 0.7827 Epoch 200/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8113 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 201/3000 1712/1712 [==============================] - 1s - loss: 0.0016 - acc: 0.7815 - val_loss: 0.0023 - val_acc: 0.7500 Epoch 202/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8026 - val_loss: 0.0024 - val_acc: 0.7780 Epoch 203/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8119 - val_loss: 0.0018 - val_acc: 0.7313 Epoch 204/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.7973 - val_loss: 0.0017 - val_acc: 0.7196 Epoch 205/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7985 - val_loss: 0.0019 - val_acc: 0.7827 Epoch 206/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7938 - val_loss: 0.0015 - val_acc: 0.7734 Epoch 207/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8078 - val_loss: 0.0021 - val_acc: 0.7266 Epoch 208/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.7996 - val_loss: 0.0021 - val_acc: 0.7430 Epoch 209/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.8002 - val_loss: 0.0020 - val_acc: 0.7710 Epoch 210/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8020 - val_loss: 0.0029 - val_acc: 0.7009 Epoch 211/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7979 - val_loss: 0.0015 - val_acc: 0.7336 Epoch 212/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.8043 - val_loss: 0.0022 - val_acc: 0.7266 Epoch 213/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.7985 - val_loss: 0.0017 - val_acc: 0.7897 Epoch 214/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7932 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 215/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8189 - val_loss: 0.0018 - val_acc: 0.7874 Epoch 216/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7856 - val_loss: 0.0014 - val_acc: 0.7617 Epoch 217/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8072 - val_loss: 0.0017 - val_acc: 0.7734 Epoch 218/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7985 - val_loss: 0.0018 - val_acc: 0.7710 Epoch 219/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8137 - val_loss: 0.0019 - val_acc: 0.7407 Epoch 220/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8043 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 221/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7961 - val_loss: 0.0030 - val_acc: 0.6963 Epoch 222/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.7973 - val_loss: 0.0015 - val_acc: 0.7196 Epoch 223/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8037 - val_loss: 0.0027 - val_acc: 0.7033 Epoch 224/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8014 - val_loss: 0.0020 - val_acc: 0.7664 Epoch 225/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.7967 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 226/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8119 - val_loss: 0.0025 - val_acc: 0.6986 Epoch 227/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8107 - val_loss: 0.0024 - val_acc: 0.7173 Epoch 228/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8014 - val_loss: 0.0023 - val_acc: 0.7079 Epoch 229/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.7956 - val_loss: 0.0018 - val_acc: 0.7453 Epoch 230/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8078 - val_loss: 0.0025 - val_acc: 0.7103 Epoch 231/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8037 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 232/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.7926 - val_loss: 0.0013 - val_acc: 0.8107 Epoch 233/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.7967 - val_loss: 0.0018 - val_acc: 0.7710 Epoch 234/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.7961 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 235/3000 1712/1712 [==============================] - 1s - loss: 0.0014 - acc: 0.8078 - val_loss: 0.0018 - val_acc: 0.7196 Epoch 236/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8166 - val_loss: 0.0019 - val_acc: 0.7430 Epoch 237/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8026 - val_loss: 0.0015 - val_acc: 0.8154 Epoch 238/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8014 - val_loss: 0.0014 - val_acc: 0.7944 Epoch 239/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.7985 - val_loss: 0.0020 - val_acc: 0.7570 Epoch 240/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8131 - val_loss: 0.0021 - val_acc: 0.7266 Epoch 241/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8166 - val_loss: 0.0015 - val_acc: 0.7453 Epoch 242/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8119 - val_loss: 0.0019 - val_acc: 0.7523 Epoch 243/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8254 - val_loss: 0.0019 - val_acc: 0.7593 Epoch 244/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.7973 - val_loss: 0.0018 - val_acc: 0.7266 Epoch 245/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8113 - val_loss: 0.0016 - val_acc: 0.7547 Epoch 246/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8026 - val_loss: 0.0023 - val_acc: 0.7336 Epoch 247/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.7967 - val_loss: 0.0015 - val_acc: 0.7477 Epoch 248/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.7932 - val_loss: 0.0023 - val_acc: 0.7336 Epoch 249/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8137 - val_loss: 0.0014 - val_acc: 0.7500 Epoch 250/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 251/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.7909 - val_loss: 0.0016 - val_acc: 0.7336 Epoch 252/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8137 - val_loss: 0.0021 - val_acc: 0.7570 Epoch 253/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8107 - val_loss: 0.0014 - val_acc: 0.7383 Epoch 254/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.7915 - val_loss: 0.0029 - val_acc: 0.7173 Epoch 255/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8084 - val_loss: 0.0021 - val_acc: 0.7477 Epoch 256/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8049 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 257/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8119 - val_loss: 0.0016 - val_acc: 0.7710 Epoch 258/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8236 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 259/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8107 - val_loss: 0.0014 - val_acc: 0.7734 Epoch 260/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.7950 - val_loss: 0.0017 - val_acc: 0.7593 Epoch 261/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8113 - val_loss: 0.0016 - val_acc: 0.8037 Epoch 262/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8242 - val_loss: 0.0015 - val_acc: 0.7664 Epoch 263/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8055 - val_loss: 0.0016 - val_acc: 0.7827 Epoch 264/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8096 - val_loss: 0.0016 - val_acc: 0.7196 Epoch 265/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8032 - val_loss: 0.0014 - val_acc: 0.7453 Epoch 266/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8265 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 267/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.7915 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 268/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8230 - val_loss: 0.0015 - val_acc: 0.7827 Epoch 269/3000 1712/1712 [==============================] - 1s - loss: 0.0013 - acc: 0.8067 - val_loss: 0.0023 - val_acc: 0.7009 Epoch 270/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8236 - val_loss: 0.0020 - val_acc: 0.7407 Epoch 271/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8078 - val_loss: 0.0024 - val_acc: 0.7313 Epoch 272/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8178 - val_loss: 0.0025 - val_acc: 0.7173 Epoch 273/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8072 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 274/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8084 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 275/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8166 - val_loss: 0.0012 - val_acc: 0.8178 Epoch 276/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8119 - val_loss: 0.0021 - val_acc: 0.7874 Epoch 277/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8067 - val_loss: 0.0021 - val_acc: 0.7173 Epoch 278/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8096 - val_loss: 0.0014 - val_acc: 0.7547 Epoch 279/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8043 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 280/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8131 - val_loss: 0.0026 - val_acc: 0.7150 Epoch 281/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8002 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 282/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8166 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 283/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8014 - val_loss: 0.0014 - val_acc: 0.7570 Epoch 284/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8143 - val_loss: 0.0019 - val_acc: 0.7126 Epoch 285/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8137 - val_loss: 0.0027 - val_acc: 0.7243 Epoch 286/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8143 - val_loss: 0.0015 - val_acc: 0.8014 Epoch 287/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8230 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 288/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8160 - val_loss: 0.0018 - val_acc: 0.7617 Epoch 289/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8172 - val_loss: 0.0016 - val_acc: 0.7477 Epoch 290/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8096 - val_loss: 0.0015 - val_acc: 0.7897 Epoch 291/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8154 - val_loss: 0.0020 - val_acc: 0.7266 Epoch 292/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8154 - val_loss: 0.0017 - val_acc: 0.7500 Epoch 293/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8067 - val_loss: 0.0015 - val_acc: 0.7640 Epoch 294/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8125 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 295/3000 1712/1712 [==============================] - 1s - loss: 9.9881e-04 - acc: 0.8166 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 296/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8154 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 297/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8102 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 298/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8218 - val_loss: 0.0015 - val_acc: 0.7874 Epoch 299/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8055 - val_loss: 0.0018 - val_acc: 0.7150 Epoch 300/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8008 - val_loss: 0.0030 - val_acc: 0.7220 Epoch 301/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8113 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 302/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8090 - val_loss: 0.0024 - val_acc: 0.7243 Epoch 303/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8113 - val_loss: 0.0018 - val_acc: 0.8014 Epoch 304/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8218 - val_loss: 0.0017 - val_acc: 0.7407 Epoch 305/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8259 - val_loss: 0.0014 - val_acc: 0.8107 Epoch 306/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8166 - val_loss: 0.0019 - val_acc: 0.7196 Epoch 307/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.7979 - val_loss: 0.0017 - val_acc: 0.7593 Epoch 308/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8201 - val_loss: 0.0015 - val_acc: 0.7640 Epoch 309/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8189 - val_loss: 0.0019 - val_acc: 0.7243 Epoch 310/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8067 - val_loss: 0.0024 - val_acc: 0.7266 Epoch 311/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8189 - val_loss: 0.0019 - val_acc: 0.7173 Epoch 312/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.7944 - val_loss: 0.0019 - val_acc: 0.7640 Epoch 313/3000 1712/1712 [==============================] - 1s - loss: 9.9185e-04 - acc: 0.8329 - val_loss: 0.0016 - val_acc: 0.7827 Epoch 314/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8125 - val_loss: 0.0014 - val_acc: 0.7944 Epoch 315/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 316/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8113 - val_loss: 0.0025 - val_acc: 0.7360 Epoch 317/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8119 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 318/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8172 - val_loss: 0.0018 - val_acc: 0.7687 Epoch 319/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8148 - val_loss: 0.0019 - val_acc: 0.7757 Epoch 320/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8113 - val_loss: 0.0018 - val_acc: 0.7617 Epoch 321/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8055 - val_loss: 0.0017 - val_acc: 0.7757 Epoch 322/3000 1712/1712 [==============================] - 1s - loss: 0.0012 - acc: 0.8125 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 323/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 324/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8224 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 325/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8090 - val_loss: 0.0014 - val_acc: 0.7500 Epoch 326/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8148 - val_loss: 0.0022 - val_acc: 0.7710 Epoch 327/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 328/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8143 - val_loss: 0.0020 - val_acc: 0.7407 Epoch 329/3000 1712/1712 [==============================] - 1s - loss: 9.9032e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 330/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8090 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 331/3000 1712/1712 [==============================] - 1s - loss: 9.9806e-04 - acc: 0.8236 - val_loss: 0.0015 - val_acc: 0.7897 Epoch 332/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8125 - val_loss: 0.0019 - val_acc: 0.6986 Epoch 333/3000 1712/1712 [==============================] - 1s - loss: 9.9615e-04 - acc: 0.8131 - val_loss: 0.0023 - val_acc: 0.7009 Epoch 334/3000 1712/1712 [==============================] - 1s - loss: 9.7830e-04 - acc: 0.8107 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 335/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8113 - val_loss: 0.0020 - val_acc: 0.7266 Epoch 336/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8166 - val_loss: 0.0022 - val_acc: 0.7593 Epoch 337/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8183 - val_loss: 0.0018 - val_acc: 0.7687 Epoch 338/3000 1712/1712 [==============================] - 1s - loss: 9.7420e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 339/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8259 - val_loss: 0.0012 - val_acc: 0.8248 Epoch 340/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8020 - val_loss: 0.0021 - val_acc: 0.7173 Epoch 341/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8160 - val_loss: 0.0017 - val_acc: 0.7664 Epoch 342/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8195 - val_loss: 0.0014 - val_acc: 0.8131 Epoch 343/3000 1712/1712 [==============================] - 1s - loss: 9.9499e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 344/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8078 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 345/3000 1712/1712 [==============================] - 1s - loss: 9.9111e-04 - acc: 0.8008 - val_loss: 0.0016 - val_acc: 0.7570 Epoch 346/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8183 - val_loss: 0.0021 - val_acc: 0.7430 Epoch 347/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8102 - val_loss: 0.0018 - val_acc: 0.7710 Epoch 348/3000 1712/1712 [==============================] - 1s - loss: 9.4751e-04 - acc: 0.8236 - val_loss: 0.0020 - val_acc: 0.7126 Epoch 349/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8032 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 350/3000 1712/1712 [==============================] - 1s - loss: 9.4011e-04 - acc: 0.8143 - val_loss: 0.0019 - val_acc: 0.7523 Epoch 351/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 352/3000 1712/1712 [==============================] - 1s - loss: 9.9234e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 353/3000 1712/1712 [==============================] - 1s - loss: 9.6271e-04 - acc: 0.8242 - val_loss: 0.0017 - val_acc: 0.8061 Epoch 354/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8242 - val_loss: 0.0014 - val_acc: 0.7991 Epoch 355/3000 1712/1712 [==============================] - 1s - loss: 9.3389e-04 - acc: 0.8201 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 356/3000 1712/1712 [==============================] - 1s - loss: 9.9948e-04 - acc: 0.8084 - val_loss: 0.0014 - val_acc: 0.7850 Epoch 357/3000 1712/1712 [==============================] - 1s - loss: 9.0407e-04 - acc: 0.8213 - val_loss: 0.0023 - val_acc: 0.7921 Epoch 358/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8131 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 359/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8207 - val_loss: 0.0014 - val_acc: 0.7757 Epoch 360/3000 1712/1712 [==============================] - 1s - loss: 8.8394e-04 - acc: 0.8294 - val_loss: 0.0020 - val_acc: 0.7523 Epoch 361/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.7996 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 362/3000 1712/1712 [==============================] - 1s - loss: 9.9608e-04 - acc: 0.8195 - val_loss: 0.0016 - val_acc: 0.8037 Epoch 363/3000 1712/1712 [==============================] - 1s - loss: 9.3238e-04 - acc: 0.8172 - val_loss: 0.0014 - val_acc: 0.8154 Epoch 364/3000 1712/1712 [==============================] - 1s - loss: 9.5899e-04 - acc: 0.8254 - val_loss: 0.0021 - val_acc: 0.7453 Epoch 365/3000 1712/1712 [==============================] - 1s - loss: 9.5775e-04 - acc: 0.8207 - val_loss: 0.0016 - val_acc: 0.7827 Epoch 366/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8113 - val_loss: 0.0019 - val_acc: 0.7407 Epoch 367/3000 1712/1712 [==============================] - 1s - loss: 9.7501e-04 - acc: 0.8207 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 368/3000 1712/1712 [==============================] - 1s - loss: 8.9631e-04 - acc: 0.8224 - val_loss: 0.0020 - val_acc: 0.7804 Epoch 369/3000 1712/1712 [==============================] - 1s - loss: 9.8787e-04 - acc: 0.8318 - val_loss: 0.0018 - val_acc: 0.7780 Epoch 370/3000 1712/1712 [==============================] - 1s - loss: 9.4248e-04 - acc: 0.8113 - val_loss: 0.0014 - val_acc: 0.7734 Epoch 371/3000 1712/1712 [==============================] - 1s - loss: 9.6018e-04 - acc: 0.8201 - val_loss: 0.0014 - val_acc: 0.8037 Epoch 372/3000 1712/1712 [==============================] - 1s - loss: 8.7642e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 373/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8201 - val_loss: 0.0014 - val_acc: 0.7593 Epoch 374/3000 1712/1712 [==============================] - 1s - loss: 8.9697e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 375/3000 1712/1712 [==============================] - 1s - loss: 9.7500e-04 - acc: 0.8230 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 376/3000 1712/1712 [==============================] - 1s - loss: 9.0987e-04 - acc: 0.8166 - val_loss: 0.0019 - val_acc: 0.7687 Epoch 377/3000 1712/1712 [==============================] - 1s - loss: 9.6591e-04 - acc: 0.8084 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 378/3000 1712/1712 [==============================] - 1s - loss: 9.7788e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 379/3000 1712/1712 [==============================] - 1s - loss: 8.8390e-04 - acc: 0.8201 - val_loss: 0.0017 - val_acc: 0.7734 Epoch 380/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8154 - val_loss: 0.0016 - val_acc: 0.7944 Epoch 381/3000 1712/1712 [==============================] - 1s - loss: 9.2199e-04 - acc: 0.8178 - val_loss: 0.0017 - val_acc: 0.7477 Epoch 382/3000 1712/1712 [==============================] - 1s - loss: 9.7649e-04 - acc: 0.8236 - val_loss: 0.0017 - val_acc: 0.7430 Epoch 383/3000 1712/1712 [==============================] - 1s - loss: 9.5232e-04 - acc: 0.8218 - val_loss: 0.0022 - val_acc: 0.7523 Epoch 384/3000 1712/1712 [==============================] - 1s - loss: 8.7891e-04 - acc: 0.8318 - val_loss: 0.0027 - val_acc: 0.7266 Epoch 385/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8341 - val_loss: 0.0017 - val_acc: 0.7687 Epoch 386/3000 1712/1712 [==============================] - 1s - loss: 9.5633e-04 - acc: 0.8218 - val_loss: 0.0021 - val_acc: 0.7150 Epoch 387/3000 1712/1712 [==============================] - 1s - loss: 9.8674e-04 - acc: 0.8043 - val_loss: 0.0017 - val_acc: 0.7150 Epoch 388/3000 1712/1712 [==============================] - 1s - loss: 9.7717e-04 - acc: 0.8090 - val_loss: 0.0017 - val_acc: 0.7734 Epoch 389/3000 1712/1712 [==============================] - 1s - loss: 9.5188e-04 - acc: 0.8125 - val_loss: 0.0020 - val_acc: 0.7897 Epoch 390/3000 1712/1712 [==============================] - 1s - loss: 9.6016e-04 - acc: 0.8341 - val_loss: 0.0018 - val_acc: 0.7570 Epoch 391/3000 1712/1712 [==============================] - 1s - loss: 9.0855e-04 - acc: 0.8224 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 392/3000 1712/1712 [==============================] - 1s - loss: 9.1827e-04 - acc: 0.8201 - val_loss: 0.0012 - val_acc: 0.8154 Epoch 393/3000 1712/1712 [==============================] - 1s - loss: 9.0780e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 394/3000 1712/1712 [==============================] - 1s - loss: 9.7460e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 395/3000 1712/1712 [==============================] - 1s - loss: 9.2267e-04 - acc: 0.8300 - val_loss: 0.0023 - val_acc: 0.6963 Epoch 396/3000 1712/1712 [==============================] - 1s - loss: 9.4010e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 397/3000 1712/1712 [==============================] - 1s - loss: 9.2698e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 398/3000 1712/1712 [==============================] - 1s - loss: 9.9163e-04 - acc: 0.8055 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 399/3000 1712/1712 [==============================] - 1s - loss: 9.5075e-04 - acc: 0.8224 - val_loss: 0.0016 - val_acc: 0.7453 Epoch 400/3000 1712/1712 [==============================] - 1s - loss: 8.7926e-04 - acc: 0.8359 - val_loss: 0.0017 - val_acc: 0.7921 Epoch 401/3000 1712/1712 [==============================] - 1s - loss: 9.7009e-04 - acc: 0.8236 - val_loss: 0.0019 - val_acc: 0.7407 Epoch 402/3000 1712/1712 [==============================] - 1s - loss: 9.1815e-04 - acc: 0.8324 - val_loss: 0.0015 - val_acc: 0.7664 Epoch 403/3000 1712/1712 [==============================] - 1s - loss: 9.6913e-04 - acc: 0.8207 - val_loss: 0.0015 - val_acc: 0.7991 Epoch 404/3000 1712/1712 [==============================] - 1s - loss: 8.6381e-04 - acc: 0.8248 - val_loss: 0.0015 - val_acc: 0.7991 Epoch 405/3000 1712/1712 [==============================] - 1s - loss: 9.4841e-04 - acc: 0.8207 - val_loss: 0.0014 - val_acc: 0.7991 Epoch 406/3000 1712/1712 [==============================] - 1s - loss: 9.0620e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.8201 Epoch 407/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 408/3000 1712/1712 [==============================] - 1s - loss: 9.3193e-04 - acc: 0.8464 - val_loss: 0.0017 - val_acc: 0.7547 Epoch 409/3000 1712/1712 [==============================] - 1s - loss: 9.1775e-04 - acc: 0.8277 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 410/3000 1712/1712 [==============================] - 1s - loss: 9.8063e-04 - acc: 0.8143 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 411/3000 1712/1712 [==============================] - 1s - loss: 9.4473e-04 - acc: 0.8113 - val_loss: 0.0015 - val_acc: 0.7664 Epoch 412/3000 1712/1712 [==============================] - 1s - loss: 8.8530e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.8131 Epoch 413/3000 1712/1712 [==============================] - 1s - loss: 9.3708e-04 - acc: 0.8137 - val_loss: 0.0012 - val_acc: 0.7617 Epoch 414/3000 1712/1712 [==============================] - 1s - loss: 9.4104e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.8224 Epoch 415/3000 1712/1712 [==============================] - 1s - loss: 9.5279e-04 - acc: 0.8195 - val_loss: 0.0014 - val_acc: 0.8037 Epoch 416/3000 1712/1712 [==============================] - 1s - loss: 8.7447e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 417/3000 1712/1712 [==============================] - 1s - loss: 9.2678e-04 - acc: 0.8102 - val_loss: 0.0013 - val_acc: 0.8178 Epoch 418/3000 1712/1712 [==============================] - 1s - loss: 8.9675e-04 - acc: 0.8137 - val_loss: 0.0021 - val_acc: 0.7757 Epoch 419/3000 1712/1712 [==============================] - 1s - loss: 9.5492e-04 - acc: 0.8248 - val_loss: 0.0017 - val_acc: 0.7757 Epoch 420/3000 1712/1712 [==============================] - 1s - loss: 8.5300e-04 - acc: 0.8341 - val_loss: 0.0019 - val_acc: 0.7523 Epoch 421/3000 1712/1712 [==============================] - 1s - loss: 9.5258e-04 - acc: 0.8189 - val_loss: 0.0015 - val_acc: 0.7944 Epoch 422/3000 1712/1712 [==============================] - 1s - loss: 8.8149e-04 - acc: 0.8265 - val_loss: 0.0019 - val_acc: 0.7196 Epoch 423/3000 1712/1712 [==============================] - 1s - loss: 9.6542e-04 - acc: 0.8254 - val_loss: 0.0019 - val_acc: 0.7196 Epoch 424/3000 1712/1712 [==============================] - 1s - loss: 8.8213e-04 - acc: 0.8353 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 425/3000 1712/1712 [==============================] - 1s - loss: 8.5027e-04 - acc: 0.8300 - val_loss: 0.0017 - val_acc: 0.7664 Epoch 426/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8230 - val_loss: 0.0015 - val_acc: 0.7523 Epoch 427/3000 1712/1712 [==============================] - 1s - loss: 8.5742e-04 - acc: 0.8335 - val_loss: 0.0014 - val_acc: 0.7967 Epoch 428/3000 1712/1712 [==============================] - 1s - loss: 9.2571e-04 - acc: 0.8189 - val_loss: 0.0014 - val_acc: 0.7967 Epoch 429/3000 1712/1712 [==============================] - 1s - loss: 8.8025e-04 - acc: 0.8300 - val_loss: 0.0018 - val_acc: 0.7827 Epoch 430/3000 1712/1712 [==============================] - 1s - loss: 9.4521e-04 - acc: 0.8154 - val_loss: 0.0019 - val_acc: 0.7243 Epoch 431/3000 1712/1712 [==============================] - 1s - loss: 8.5353e-04 - acc: 0.8254 - val_loss: 0.0020 - val_acc: 0.7290 Epoch 432/3000 1712/1712 [==============================] - 1s - loss: 8.8619e-04 - acc: 0.8300 - val_loss: 0.0018 - val_acc: 0.7617 Epoch 433/3000 1712/1712 [==============================] - 1s - loss: 9.4204e-04 - acc: 0.8090 - val_loss: 0.0021 - val_acc: 0.7360 Epoch 434/3000 1712/1712 [==============================] - 1s - loss: 8.7698e-04 - acc: 0.8213 - val_loss: 0.0021 - val_acc: 0.7430 Epoch 435/3000 1712/1712 [==============================] - 1s - loss: 9.3837e-04 - acc: 0.8259 - val_loss: 0.0015 - val_acc: 0.7921 Epoch 436/3000 1712/1712 [==============================] - 1s - loss: 8.3513e-04 - acc: 0.8318 - val_loss: 0.0027 - val_acc: 0.7033 Epoch 437/3000 1712/1712 [==============================] - 1s - loss: 9.3110e-04 - acc: 0.8318 - val_loss: 0.0011 - val_acc: 0.8248 Epoch 438/3000 1712/1712 [==============================] - 1s - loss: 9.4539e-04 - acc: 0.7979 - val_loss: 0.0014 - val_acc: 0.7196 Epoch 439/3000 1712/1712 [==============================] - 1s - loss: 8.5315e-04 - acc: 0.8189 - val_loss: 0.0014 - val_acc: 0.8014 Epoch 440/3000 1712/1712 [==============================] - 1s - loss: 0.0011 - acc: 0.8172 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 441/3000 1712/1712 [==============================] - 1s - loss: 8.4199e-04 - acc: 0.8382 - val_loss: 0.0019 - val_acc: 0.7126 Epoch 442/3000 1712/1712 [==============================] - 1s - loss: 8.9495e-04 - acc: 0.8318 - val_loss: 0.0016 - val_acc: 0.7640 Epoch 443/3000 1712/1712 [==============================] - 1s - loss: 8.9638e-04 - acc: 0.8183 - val_loss: 0.0016 - val_acc: 0.7804 Epoch 444/3000 1712/1712 [==============================] - 1s - loss: 8.4086e-04 - acc: 0.8183 - val_loss: 0.0025 - val_acc: 0.7103 Epoch 445/3000 1712/1712 [==============================] - 1s - loss: 8.6085e-04 - acc: 0.8213 - val_loss: 0.0017 - val_acc: 0.7500 Epoch 446/3000 1712/1712 [==============================] - 1s - loss: 9.4103e-04 - acc: 0.8154 - val_loss: 0.0016 - val_acc: 0.7336 Epoch 447/3000 1712/1712 [==============================] - 1s - loss: 9.8905e-04 - acc: 0.8102 - val_loss: 0.0016 - val_acc: 0.7991 Epoch 448/3000 1712/1712 [==============================] - 1s - loss: 7.9793e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.8154 Epoch 449/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 450/3000 1712/1712 [==============================] - 1s - loss: 7.8798e-04 - acc: 0.8236 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 451/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8207 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 452/3000 1712/1712 [==============================] - 1s - loss: 8.6245e-04 - acc: 0.8271 - val_loss: 0.0017 - val_acc: 0.7407 Epoch 453/3000 1712/1712 [==============================] - 1s - loss: 9.0492e-04 - acc: 0.8318 - val_loss: 0.0014 - val_acc: 0.8084 Epoch 454/3000 1712/1712 [==============================] - 1s - loss: 8.1789e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 455/3000 1712/1712 [==============================] - 1s - loss: 8.5080e-04 - acc: 0.8329 - val_loss: 0.0023 - val_acc: 0.7313 Epoch 456/3000 1712/1712 [==============================] - 1s - loss: 9.7834e-04 - acc: 0.8201 - val_loss: 0.0017 - val_acc: 0.7407 Epoch 457/3000 1712/1712 [==============================] - 1s - loss: 9.0112e-04 - acc: 0.8090 - val_loss: 0.0020 - val_acc: 0.7360 Epoch 458/3000 1712/1712 [==============================] - 1s - loss: 8.4896e-04 - acc: 0.8189 - val_loss: 0.0015 - val_acc: 0.7944 Epoch 459/3000 1712/1712 [==============================] - 1s - loss: 9.8994e-04 - acc: 0.8283 - val_loss: 0.0015 - val_acc: 0.7921 Epoch 460/3000 1712/1712 [==============================] - 1s - loss: 8.0593e-04 - acc: 0.8160 - val_loss: 0.0021 - val_acc: 0.7850 Epoch 461/3000 1712/1712 [==============================] - 1s - loss: 9.4345e-04 - acc: 0.8102 - val_loss: 0.0017 - val_acc: 0.7360 Epoch 462/3000 1712/1712 [==============================] - 1s - loss: 8.3462e-04 - acc: 0.8254 - val_loss: 0.0021 - val_acc: 0.7150 Epoch 463/3000 1712/1712 [==============================] - 1s - loss: 9.4450e-04 - acc: 0.8213 - val_loss: 0.0014 - val_acc: 0.7967 Epoch 464/3000 1712/1712 [==============================] - 1s - loss: 9.0350e-04 - acc: 0.8341 - val_loss: 0.0019 - val_acc: 0.7640 Epoch 465/3000 1712/1712 [==============================] - 1s - loss: 8.4225e-04 - acc: 0.8359 - val_loss: 0.0017 - val_acc: 0.7430 Epoch 466/3000 1712/1712 [==============================] - 1s - loss: 9.5289e-04 - acc: 0.8248 - val_loss: 0.0015 - val_acc: 0.7430 Epoch 467/3000 1712/1712 [==============================] - 1s - loss: 8.9614e-04 - acc: 0.8213 - val_loss: 0.0015 - val_acc: 0.7640 Epoch 468/3000 1712/1712 [==============================] - 1s - loss: 9.5627e-04 - acc: 0.8131 - val_loss: 0.0018 - val_acc: 0.7710 Epoch 469/3000 1712/1712 [==============================] - 1s - loss: 8.1707e-04 - acc: 0.8283 - val_loss: 0.0015 - val_acc: 0.7921 Epoch 470/3000 1712/1712 [==============================] - 1s - loss: 8.6089e-04 - acc: 0.8254 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 471/3000 1712/1712 [==============================] - 1s - loss: 9.1114e-04 - acc: 0.8213 - val_loss: 0.0014 - val_acc: 0.8154 Epoch 472/3000 1712/1712 [==============================] - 1s - loss: 8.9131e-04 - acc: 0.8400 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 473/3000 1712/1712 [==============================] - 1s - loss: 7.7119e-04 - acc: 0.8242 - val_loss: 0.0014 - val_acc: 0.8107 Epoch 474/3000 1712/1712 [==============================] - 1s - loss: 9.4479e-04 - acc: 0.8061 - val_loss: 0.0022 - val_acc: 0.7290 Epoch 475/3000 1712/1712 [==============================] - 1s - loss: 9.2586e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.8154 Epoch 476/3000 1712/1712 [==============================] - 1s - loss: 8.8425e-04 - acc: 0.8230 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 477/3000 1712/1712 [==============================] - 1s - loss: 9.1195e-04 - acc: 0.8125 - val_loss: 0.0014 - val_acc: 0.8388 Epoch 478/3000 1712/1712 [==============================] - 1s - loss: 8.0246e-04 - acc: 0.8306 - val_loss: 0.0025 - val_acc: 0.7243 Epoch 479/3000 1712/1712 [==============================] - 1s - loss: 9.8900e-04 - acc: 0.8119 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 480/3000 1712/1712 [==============================] - 1s - loss: 9.1234e-04 - acc: 0.8218 - val_loss: 0.0020 - val_acc: 0.7150 Epoch 481/3000 1712/1712 [==============================] - 1s - loss: 8.1241e-04 - acc: 0.8300 - val_loss: 0.0023 - val_acc: 0.7266 Epoch 482/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 483/3000 1712/1712 [==============================] - 1s - loss: 8.4058e-04 - acc: 0.8189 - val_loss: 0.0011 - val_acc: 0.8037 Epoch 484/3000 1712/1712 [==============================] - 1s - loss: 9.2092e-04 - acc: 0.8218 - val_loss: 0.0015 - val_acc: 0.8107 Epoch 485/3000 1712/1712 [==============================] - 1s - loss: 8.9424e-04 - acc: 0.8277 - val_loss: 0.0011 - val_acc: 0.8248 Epoch 486/3000 1712/1712 [==============================] - 1s - loss: 8.4825e-04 - acc: 0.8230 - val_loss: 0.0014 - val_acc: 0.7944 Epoch 487/3000 1712/1712 [==============================] - 1s - loss: 9.3373e-04 - acc: 0.8283 - val_loss: 0.0019 - val_acc: 0.7500 Epoch 488/3000 1712/1712 [==============================] - 1s - loss: 8.2050e-04 - acc: 0.8236 - val_loss: 0.0022 - val_acc: 0.7336 Epoch 489/3000 1712/1712 [==============================] - 1s - loss: 9.6247e-04 - acc: 0.8242 - val_loss: 0.0019 - val_acc: 0.7804 Epoch 490/3000 1712/1712 [==============================] - 1s - loss: 8.6195e-04 - acc: 0.8189 - val_loss: 0.0018 - val_acc: 0.7804 Epoch 491/3000 1712/1712 [==============================] - 1s - loss: 8.6671e-04 - acc: 0.8113 - val_loss: 0.0023 - val_acc: 0.7313 Epoch 492/3000 1712/1712 [==============================] - 1s - loss: 8.8587e-04 - acc: 0.8125 - val_loss: 0.0017 - val_acc: 0.7407 Epoch 493/3000 1712/1712 [==============================] - 1s - loss: 9.0602e-04 - acc: 0.8283 - val_loss: 0.0016 - val_acc: 0.7710 Epoch 494/3000 1712/1712 [==============================] - 1s - loss: 9.1215e-04 - acc: 0.8248 - val_loss: 0.0018 - val_acc: 0.7944 Epoch 495/3000 1712/1712 [==============================] - 1s - loss: 8.3522e-04 - acc: 0.8364 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 496/3000 1712/1712 [==============================] - 1s - loss: 9.5850e-04 - acc: 0.8230 - val_loss: 0.0012 - val_acc: 0.8248 Epoch 497/3000 1712/1712 [==============================] - 1s - loss: 8.5328e-04 - acc: 0.8224 - val_loss: 0.0014 - val_acc: 0.7570 Epoch 498/3000 1712/1712 [==============================] - 1s - loss: 8.7950e-04 - acc: 0.8236 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 499/3000 1712/1712 [==============================] - 1s - loss: 8.6498e-04 - acc: 0.8084 - val_loss: 0.0019 - val_acc: 0.8107 Epoch 500/3000 1712/1712 [==============================] - 1s - loss: 8.4509e-04 - acc: 0.8429 - val_loss: 0.0012 - val_acc: 0.8201 Epoch 501/3000 1712/1712 [==============================] - 1s - loss: 8.8951e-04 - acc: 0.8230 - val_loss: 0.0012 - val_acc: 0.8201 Epoch 502/3000 1712/1712 [==============================] - 1s - loss: 8.7373e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.8248 Epoch 503/3000 1712/1712 [==============================] - 1s - loss: 8.8398e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 504/3000 1712/1712 [==============================] - 1s - loss: 8.5853e-04 - acc: 0.8224 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 505/3000 1712/1712 [==============================] - 1s - loss: 9.3320e-04 - acc: 0.8090 - val_loss: 0.0015 - val_acc: 0.7921 Epoch 506/3000 1712/1712 [==============================] - 1s - loss: 8.5504e-04 - acc: 0.8283 - val_loss: 0.0015 - val_acc: 0.7617 Epoch 507/3000 1712/1712 [==============================] - 1s - loss: 7.8426e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.8061 Epoch 508/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8254 - val_loss: 0.0014 - val_acc: 0.7967 Epoch 509/3000 1712/1712 [==============================] - 1s - loss: 8.2601e-04 - acc: 0.8277 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 510/3000 1712/1712 [==============================] - 1s - loss: 8.4423e-04 - acc: 0.8318 - val_loss: 0.0011 - val_acc: 0.8154 Epoch 511/3000 1712/1712 [==============================] - 1s - loss: 8.6867e-04 - acc: 0.8218 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 512/3000 1712/1712 [==============================] - 1s - loss: 8.9835e-04 - acc: 0.8464 - val_loss: 0.0015 - val_acc: 0.8107 Epoch 513/3000 1712/1712 [==============================] - 1s - loss: 9.7207e-04 - acc: 0.8178 - val_loss: 0.0017 - val_acc: 0.7383 Epoch 514/3000 1712/1712 [==============================] - 1s - loss: 8.4765e-04 - acc: 0.8254 - val_loss: 0.0019 - val_acc: 0.7383 Epoch 515/3000 1712/1712 [==============================] - 1s - loss: 8.5412e-04 - acc: 0.8335 - val_loss: 0.0019 - val_acc: 0.7687 Epoch 516/3000 1712/1712 [==============================] - 1s - loss: 9.4626e-04 - acc: 0.8201 - val_loss: 0.0021 - val_acc: 0.7383 Epoch 517/3000 1712/1712 [==============================] - 1s - loss: 8.4366e-04 - acc: 0.8347 - val_loss: 0.0019 - val_acc: 0.7383 Epoch 518/3000 1712/1712 [==============================] - 1s - loss: 9.0261e-04 - acc: 0.8370 - val_loss: 0.0017 - val_acc: 0.7570 Epoch 519/3000 1712/1712 [==============================] - 1s - loss: 8.4425e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 520/3000 1712/1712 [==============================] - 1s - loss: 8.9676e-04 - acc: 0.8207 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 521/3000 1712/1712 [==============================] - 1s - loss: 8.4157e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 522/3000 1712/1712 [==============================] - 1s - loss: 9.2305e-04 - acc: 0.8183 - val_loss: 0.0013 - val_acc: 0.8154 Epoch 523/3000 1712/1712 [==============================] - 1s - loss: 8.2170e-04 - acc: 0.8329 - val_loss: 0.0014 - val_acc: 0.8061 Epoch 524/3000 1712/1712 [==============================] - 1s - loss: 9.1653e-04 - acc: 0.8183 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 525/3000 1712/1712 [==============================] - 1s - loss: 8.5042e-04 - acc: 0.8359 - val_loss: 0.0022 - val_acc: 0.7266 Epoch 526/3000 1712/1712 [==============================] - 1s - loss: 8.5110e-04 - acc: 0.8271 - val_loss: 0.0016 - val_acc: 0.7780 Epoch 527/3000 1712/1712 [==============================] - 1s - loss: 8.9439e-04 - acc: 0.8248 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 528/3000 1712/1712 [==============================] - 1s - loss: 8.5762e-04 - acc: 0.8259 - val_loss: 0.0018 - val_acc: 0.7523 Epoch 529/3000 1712/1712 [==============================] - 1s - loss: 8.7493e-04 - acc: 0.8277 - val_loss: 0.0015 - val_acc: 0.8178 Epoch 530/3000 1712/1712 [==============================] - 1s - loss: 8.7228e-04 - acc: 0.8143 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 531/3000 1712/1712 [==============================] - 1s - loss: 9.1303e-04 - acc: 0.8242 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 532/3000 1712/1712 [==============================] - 1s - loss: 9.0854e-04 - acc: 0.8248 - val_loss: 0.0019 - val_acc: 0.7874 Epoch 533/3000 1712/1712 [==============================] - 1s - loss: 7.9492e-04 - acc: 0.8347 - val_loss: 0.0027 - val_acc: 0.7360 Epoch 534/3000 1712/1712 [==============================] - 1s - loss: 9.5255e-04 - acc: 0.8242 - val_loss: 0.0016 - val_acc: 0.7617 Epoch 535/3000 1712/1712 [==============================] - 1s - loss: 8.4310e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 536/3000 1712/1712 [==============================] - 1s - loss: 8.9513e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 537/3000 1712/1712 [==============================] - 1s - loss: 8.7901e-04 - acc: 0.8248 - val_loss: 0.0014 - val_acc: 0.7547 Epoch 538/3000 1712/1712 [==============================] - 1s - loss: 8.9923e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.7500 Epoch 539/3000 1712/1712 [==============================] - 1s - loss: 8.1421e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7477 Epoch 540/3000 1712/1712 [==============================] - 1s - loss: 9.0251e-04 - acc: 0.8294 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 541/3000 1712/1712 [==============================] - 1s - loss: 8.8326e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 542/3000 1712/1712 [==============================] - 1s - loss: 8.9526e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 543/3000 1712/1712 [==============================] - 1s - loss: 8.3864e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 544/3000 1712/1712 [==============================] - 1s - loss: 8.3111e-04 - acc: 0.8166 - val_loss: 0.0023 - val_acc: 0.7243 Epoch 545/3000 1712/1712 [==============================] - 1s - loss: 9.2765e-04 - acc: 0.8189 - val_loss: 0.0023 - val_acc: 0.7173 Epoch 546/3000 1712/1712 [==============================] - 1s - loss: 8.8784e-04 - acc: 0.8329 - val_loss: 0.0015 - val_acc: 0.7640 Epoch 547/3000 1712/1712 [==============================] - 1s - loss: 8.9520e-04 - acc: 0.8224 - val_loss: 0.0021 - val_acc: 0.7220 Epoch 548/3000 1712/1712 [==============================] - 1s - loss: 8.5065e-04 - acc: 0.8248 - val_loss: 0.0020 - val_acc: 0.7477 Epoch 549/3000 1712/1712 [==============================] - 1s - loss: 8.4501e-04 - acc: 0.8464 - val_loss: 0.0020 - val_acc: 0.7360 Epoch 550/3000 1712/1712 [==============================] - 1s - loss: 9.0921e-04 - acc: 0.8254 - val_loss: 0.0018 - val_acc: 0.7196 Epoch 551/3000 1712/1712 [==============================] - 1s - loss: 8.5660e-04 - acc: 0.8201 - val_loss: 0.0015 - val_acc: 0.7944 Epoch 552/3000 1712/1712 [==============================] - 1s - loss: 8.5536e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 553/3000 1712/1712 [==============================] - 1s - loss: 8.8441e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.8201 Epoch 554/3000 1712/1712 [==============================] - 1s - loss: 9.0833e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 555/3000 1712/1712 [==============================] - 1s - loss: 8.0022e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 556/3000 1712/1712 [==============================] - 1s - loss: 9.8586e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 557/3000 1712/1712 [==============================] - 1s - loss: 7.3352e-04 - acc: 0.8335 - val_loss: 0.0024 - val_acc: 0.7056 Epoch 558/3000 1712/1712 [==============================] - 1s - loss: 9.6470e-04 - acc: 0.8213 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 559/3000 1712/1712 [==============================] - 1s - loss: 7.9745e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 560/3000 1712/1712 [==============================] - 1s - loss: 9.6127e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 561/3000 1712/1712 [==============================] - 1s - loss: 8.9773e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 562/3000 1712/1712 [==============================] - 1s - loss: 7.8465e-04 - acc: 0.8324 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 563/3000 1712/1712 [==============================] - 1s - loss: 8.7582e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 564/3000 1712/1712 [==============================] - 1s - loss: 8.9618e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 565/3000 1712/1712 [==============================] - 1s - loss: 8.6771e-04 - acc: 0.8119 - val_loss: 0.0023 - val_acc: 0.7150 Epoch 566/3000 1712/1712 [==============================] - 1s - loss: 8.6077e-04 - acc: 0.8166 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 567/3000 1712/1712 [==============================] - 1s - loss: 9.2695e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 568/3000 1712/1712 [==============================] - 1s - loss: 8.0130e-04 - acc: 0.8230 - val_loss: 0.0011 - val_acc: 0.8271 Epoch 569/3000 1712/1712 [==============================] - 1s - loss: 8.3103e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 570/3000 1712/1712 [==============================] - 1s - loss: 9.0602e-04 - acc: 0.8370 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 571/3000 1712/1712 [==============================] - 1s - loss: 7.9427e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 572/3000 1712/1712 [==============================] - 1s - loss: 9.3154e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 573/3000 1712/1712 [==============================] - 1s - loss: 8.6009e-04 - acc: 0.8242 - val_loss: 0.0011 - val_acc: 0.8084 Epoch 574/3000 1712/1712 [==============================] - 1s - loss: 7.9844e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 575/3000 1712/1712 [==============================] - 1s - loss: 9.0057e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 576/3000 1712/1712 [==============================] - 1s - loss: 8.4330e-04 - acc: 0.8189 - val_loss: 0.0014 - val_acc: 0.8014 Epoch 577/3000 1712/1712 [==============================] - 1s - loss: 7.9875e-04 - acc: 0.8353 - val_loss: 0.0016 - val_acc: 0.7593 Epoch 578/3000 1712/1712 [==============================] - 1s - loss: 8.8984e-04 - acc: 0.8300 - val_loss: 0.0014 - val_acc: 0.7991 Epoch 579/3000 1712/1712 [==============================] - 1s - loss: 9.1550e-04 - acc: 0.8166 - val_loss: 0.0016 - val_acc: 0.7617 Epoch 580/3000 1712/1712 [==============================] - 1s - loss: 8.8505e-04 - acc: 0.8370 - val_loss: 0.0021 - val_acc: 0.7804 Epoch 581/3000 1712/1712 [==============================] - 1s - loss: 9.2993e-04 - acc: 0.8283 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 582/3000 1712/1712 [==============================] - 1s - loss: 8.0247e-04 - acc: 0.8312 - val_loss: 0.0021 - val_acc: 0.7874 Epoch 583/3000 1712/1712 [==============================] - 1s - loss: 9.0660e-04 - acc: 0.8411 - val_loss: 0.0020 - val_acc: 0.7523 Epoch 584/3000 1712/1712 [==============================] - 1s - loss: 8.8272e-04 - acc: 0.8283 - val_loss: 0.0015 - val_acc: 0.7453 Epoch 585/3000 1712/1712 [==============================] - 1s - loss: 8.4010e-04 - acc: 0.8370 - val_loss: 0.0021 - val_acc: 0.7266 Epoch 586/3000 1712/1712 [==============================] - 1s - loss: 8.2166e-04 - acc: 0.8224 - val_loss: 0.0013 - val_acc: 0.8178 Epoch 587/3000 1712/1712 [==============================] - 1s - loss: 9.2474e-04 - acc: 0.8026 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 588/3000 1712/1712 [==============================] - 1s - loss: 7.9778e-04 - acc: 0.8347 - val_loss: 0.0014 - val_acc: 0.8061 Epoch 589/3000 1712/1712 [==============================] - 1s - loss: 8.4633e-04 - acc: 0.8166 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 590/3000 1712/1712 [==============================] - 1s - loss: 8.1698e-04 - acc: 0.8341 - val_loss: 0.0014 - val_acc: 0.7850 Epoch 591/3000 1712/1712 [==============================] - 1s - loss: 8.6768e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 592/3000 1712/1712 [==============================] - 1s - loss: 8.8476e-04 - acc: 0.8154 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 593/3000 1712/1712 [==============================] - 1s - loss: 8.5649e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.8154 Epoch 594/3000 1712/1712 [==============================] - 1s - loss: 8.1772e-04 - acc: 0.8376 - val_loss: 0.0012 - val_acc: 0.8154 Epoch 595/3000 1712/1712 [==============================] - 1s - loss: 8.8887e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 596/3000 1712/1712 [==============================] - 1s - loss: 8.8654e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 597/3000 1712/1712 [==============================] - 1s - loss: 8.2816e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 598/3000 1712/1712 [==============================] - 1s - loss: 8.7142e-04 - acc: 0.8259 - val_loss: 0.0012 - val_acc: 0.8154 Epoch 599/3000 1712/1712 [==============================] - 1s - loss: 8.6418e-04 - acc: 0.8259 - val_loss: 0.0012 - val_acc: 0.8248 Epoch 600/3000 1712/1712 [==============================] - 1s - loss: 8.3418e-04 - acc: 0.8102 - val_loss: 0.0014 - val_acc: 0.7664 Epoch 601/3000 1712/1712 [==============================] - 1s - loss: 8.6273e-04 - acc: 0.8160 - val_loss: 0.0012 - val_acc: 0.7617 Epoch 602/3000 1712/1712 [==============================] - 1s - loss: 8.7137e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 603/3000 1712/1712 [==============================] - 1s - loss: 9.1130e-04 - acc: 0.8166 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 604/3000 1712/1712 [==============================] - 1s - loss: 8.3245e-04 - acc: 0.8277 - val_loss: 0.0011 - val_acc: 0.8201 Epoch 605/3000 1712/1712 [==============================] - 1s - loss: 8.1308e-04 - acc: 0.8364 - val_loss: 0.0011 - val_acc: 0.8014 Epoch 606/3000 1712/1712 [==============================] - 1s - loss: 8.8784e-04 - acc: 0.8224 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 607/3000 1712/1712 [==============================] - 1s - loss: 8.5112e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 608/3000 1712/1712 [==============================] - 1s - loss: 8.2753e-04 - acc: 0.8300 - val_loss: 0.0011 - val_acc: 0.8131 Epoch 609/3000 1712/1712 [==============================] - 1s - loss: 8.4903e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 610/3000 1712/1712 [==============================] - 1s - loss: 8.1233e-04 - acc: 0.8300 - val_loss: 0.0019 - val_acc: 0.7874 Epoch 611/3000 1712/1712 [==============================] - 1s - loss: 9.0337e-04 - acc: 0.8189 - val_loss: 0.0016 - val_acc: 0.7757 Epoch 612/3000 1712/1712 [==============================] - 1s - loss: 8.3554e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7477 Epoch 613/3000 1712/1712 [==============================] - 1s - loss: 8.8798e-04 - acc: 0.8230 - val_loss: 0.0016 - val_acc: 0.7664 Epoch 614/3000 1712/1712 [==============================] - 1s - loss: 8.2211e-04 - acc: 0.8283 - val_loss: 0.0020 - val_acc: 0.7196 Epoch 615/3000 1712/1712 [==============================] - 1s - loss: 8.1533e-04 - acc: 0.8359 - val_loss: 0.0011 - val_acc: 0.8084 Epoch 616/3000 1712/1712 [==============================] - 1s - loss: 9.2493e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 617/3000 1712/1712 [==============================] - 1s - loss: 8.1786e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 618/3000 1712/1712 [==============================] - 1s - loss: 8.3995e-04 - acc: 0.8259 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 619/3000 1712/1712 [==============================] - 1s - loss: 8.4430e-04 - acc: 0.8329 - val_loss: 0.0015 - val_acc: 0.7500 Epoch 620/3000 1712/1712 [==============================] - 1s - loss: 8.5922e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 621/3000 1712/1712 [==============================] - 1s - loss: 8.6352e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.7523 Epoch 622/3000 1712/1712 [==============================] - 1s - loss: 8.3675e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 623/3000 1712/1712 [==============================] - 1s - loss: 8.2059e-04 - acc: 0.8364 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 624/3000 1712/1712 [==============================] - 1s - loss: 8.7277e-04 - acc: 0.8224 - val_loss: 0.0018 - val_acc: 0.7640 Epoch 625/3000 1712/1712 [==============================] - 1s - loss: 9.3702e-04 - acc: 0.8277 - val_loss: 0.0019 - val_acc: 0.7126 Epoch 626/3000 1712/1712 [==============================] - 1s - loss: 8.3741e-04 - acc: 0.8283 - val_loss: 0.0018 - val_acc: 0.7804 Epoch 627/3000 1712/1712 [==============================] - 1s - loss: 8.7912e-04 - acc: 0.8335 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 628/3000 1712/1712 [==============================] - 1s - loss: 8.0227e-04 - acc: 0.8376 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 629/3000 1712/1712 [==============================] - 1s - loss: 9.5552e-04 - acc: 0.8230 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 630/3000 1712/1712 [==============================] - 1s - loss: 8.2016e-04 - acc: 0.8160 - val_loss: 0.0021 - val_acc: 0.7500 Epoch 631/3000 1712/1712 [==============================] - 1s - loss: 8.2509e-04 - acc: 0.8271 - val_loss: 0.0020 - val_acc: 0.7570 Epoch 632/3000 1712/1712 [==============================] - 1s - loss: 9.1822e-04 - acc: 0.8119 - val_loss: 0.0019 - val_acc: 0.7617 Epoch 633/3000 1712/1712 [==============================] - 1s - loss: 8.5147e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 634/3000 1712/1712 [==============================] - 1s - loss: 8.6582e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.7593 Epoch 635/3000 1712/1712 [==============================] - 1s - loss: 8.4677e-04 - acc: 0.8289 - val_loss: 0.0014 - val_acc: 0.8131 Epoch 636/3000 1712/1712 [==============================] - 1s - loss: 8.6761e-04 - acc: 0.8271 - val_loss: 0.0018 - val_acc: 0.7547 Epoch 637/3000 1712/1712 [==============================] - 1s - loss: 8.5100e-04 - acc: 0.8254 - val_loss: 0.0017 - val_acc: 0.7734 Epoch 638/3000 1712/1712 [==============================] - 1s - loss: 8.3106e-04 - acc: 0.8259 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 639/3000 1712/1712 [==============================] - 1s - loss: 8.7948e-04 - acc: 0.8137 - val_loss: 0.0018 - val_acc: 0.7780 Epoch 640/3000 1712/1712 [==============================] - 1s - loss: 8.7107e-04 - acc: 0.8370 - val_loss: 0.0016 - val_acc: 0.7453 Epoch 641/3000 1712/1712 [==============================] - 1s - loss: 8.2514e-04 - acc: 0.8353 - val_loss: 0.0021 - val_acc: 0.7056 Epoch 642/3000 1712/1712 [==============================] - 1s - loss: 8.7983e-04 - acc: 0.8400 - val_loss: 0.0014 - val_acc: 0.7874 Epoch 643/3000 1712/1712 [==============================] - 1s - loss: 9.3973e-04 - acc: 0.8213 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 644/3000 1712/1712 [==============================] - 1s - loss: 8.2503e-04 - acc: 0.8283 - val_loss: 0.0011 - val_acc: 0.8014 Epoch 645/3000 1712/1712 [==============================] - 1s - loss: 8.7517e-04 - acc: 0.8189 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 646/3000 1712/1712 [==============================] - 1s - loss: 8.8402e-04 - acc: 0.8400 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 647/3000 1712/1712 [==============================] - 1s - loss: 8.5504e-04 - acc: 0.8329 - val_loss: 0.0011 - val_acc: 0.8154 Epoch 648/3000 1712/1712 [==============================] - 1s - loss: 7.8661e-04 - acc: 0.8417 - val_loss: 0.0017 - val_acc: 0.7383 Epoch 649/3000 1712/1712 [==============================] - 1s - loss: 8.8995e-04 - acc: 0.8277 - val_loss: 0.0016 - val_acc: 0.7523 Epoch 650/3000 1712/1712 [==============================] - 1s - loss: 9.1363e-04 - acc: 0.8329 - val_loss: 0.0014 - val_acc: 0.8107 Epoch 651/3000 1712/1712 [==============================] - 1s - loss: 8.3007e-04 - acc: 0.8417 - val_loss: 0.0019 - val_acc: 0.7360 Epoch 652/3000 1712/1712 [==============================] - 1s - loss: 8.3798e-04 - acc: 0.8248 - val_loss: 0.0017 - val_acc: 0.7921 Epoch 653/3000 1712/1712 [==============================] - 1s - loss: 8.6969e-04 - acc: 0.8300 - val_loss: 0.0015 - val_acc: 0.7804 Epoch 654/3000 1712/1712 [==============================] - 1s - loss: 8.3067e-04 - acc: 0.8242 - val_loss: 0.0019 - val_acc: 0.7383 Epoch 655/3000 1712/1712 [==============================] - 1s - loss: 8.5844e-04 - acc: 0.8201 - val_loss: 0.0023 - val_acc: 0.7687 Epoch 656/3000 1712/1712 [==============================] - 1s - loss: 8.1486e-04 - acc: 0.8353 - val_loss: 0.0021 - val_acc: 0.7033 Epoch 657/3000 1712/1712 [==============================] - 1s - loss: 9.3098e-04 - acc: 0.8113 - val_loss: 0.0015 - val_acc: 0.7313 Epoch 658/3000 1712/1712 [==============================] - 1s - loss: 8.0899e-04 - acc: 0.8131 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 659/3000 1712/1712 [==============================] - 1s - loss: 8.4612e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 660/3000 1712/1712 [==============================] - 1s - loss: 8.4674e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 661/3000 1712/1712 [==============================] - 1s - loss: 8.6501e-04 - acc: 0.8172 - val_loss: 0.0014 - val_acc: 0.7827 Epoch 662/3000 1712/1712 [==============================] - 1s - loss: 8.9744e-04 - acc: 0.8183 - val_loss: 0.0013 - val_acc: 0.8154 Epoch 663/3000 1712/1712 [==============================] - 1s - loss: 8.4120e-04 - acc: 0.8353 - val_loss: 0.0013 - val_acc: 0.8107 Epoch 664/3000 1712/1712 [==============================] - 1s - loss: 8.4109e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 665/3000 1712/1712 [==============================] - 1s - loss: 7.7194e-04 - acc: 0.8388 - val_loss: 0.0014 - val_acc: 0.7640 Epoch 666/3000 1712/1712 [==============================] - 1s - loss: 9.4805e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.8107 Epoch 667/3000 1712/1712 [==============================] - 1s - loss: 8.0112e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 668/3000 1712/1712 [==============================] - 1s - loss: 8.4789e-04 - acc: 0.8411 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 669/3000 1712/1712 [==============================] - 1s - loss: 8.5845e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 670/3000 1712/1712 [==============================] - 1s - loss: 8.6432e-04 - acc: 0.8370 - val_loss: 0.0018 - val_acc: 0.7734 Epoch 671/3000 1712/1712 [==============================] - 1s - loss: 8.4408e-04 - acc: 0.8271 - val_loss: 0.0019 - val_acc: 0.7266 Epoch 672/3000 1712/1712 [==============================] - 1s - loss: 8.5613e-04 - acc: 0.8411 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 673/3000 1712/1712 [==============================] - 1s - loss: 8.0284e-04 - acc: 0.8277 - val_loss: 0.0011 - val_acc: 0.7944 Epoch 674/3000 1712/1712 [==============================] - 1s - loss: 8.8046e-04 - acc: 0.8300 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 675/3000 1712/1712 [==============================] - 1s - loss: 8.3984e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7477 Epoch 676/3000 1712/1712 [==============================] - 1s - loss: 8.8571e-04 - acc: 0.8172 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 677/3000 1712/1712 [==============================] - 1s - loss: 8.1809e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 678/3000 1712/1712 [==============================] - 1s - loss: 8.3311e-04 - acc: 0.8423 - val_loss: 0.0012 - val_acc: 0.8294 Epoch 679/3000 1712/1712 [==============================] - 1s - loss: 8.2376e-04 - acc: 0.8300 - val_loss: 0.0021 - val_acc: 0.7360 Epoch 680/3000 1712/1712 [==============================] - 1s - loss: 8.2904e-04 - acc: 0.8265 - val_loss: 0.0020 - val_acc: 0.7710 Epoch 681/3000 1712/1712 [==============================] - 1s - loss: 8.6173e-04 - acc: 0.8224 - val_loss: 0.0020 - val_acc: 0.7477 Epoch 682/3000 1712/1712 [==============================] - 1s - loss: 8.6829e-04 - acc: 0.8300 - val_loss: 0.0015 - val_acc: 0.8084 Epoch 683/3000 1712/1712 [==============================] - 1s - loss: 8.7833e-04 - acc: 0.8183 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 684/3000 1712/1712 [==============================] - 1s - loss: 8.0113e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 685/3000 1712/1712 [==============================] - 1s - loss: 9.2366e-04 - acc: 0.8236 - val_loss: 0.0019 - val_acc: 0.7383 Epoch 686/3000 1712/1712 [==============================] - 1s - loss: 7.8256e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 687/3000 1712/1712 [==============================] - 1s - loss: 9.1827e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 688/3000 1712/1712 [==============================] - 1s - loss: 8.2846e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.7313 Epoch 689/3000 1712/1712 [==============================] - 1s - loss: 8.3303e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 690/3000 1712/1712 [==============================] - 1s - loss: 8.2433e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 691/3000 1712/1712 [==============================] - 1s - loss: 8.3154e-04 - acc: 0.8382 - val_loss: 0.0016 - val_acc: 0.7874 Epoch 692/3000 1712/1712 [==============================] - 1s - loss: 9.0645e-04 - acc: 0.8324 - val_loss: 0.0017 - val_acc: 0.7710 Epoch 693/3000 1712/1712 [==============================] - 1s - loss: 8.0485e-04 - acc: 0.8248 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 694/3000 1712/1712 [==============================] - 1s - loss: 8.8677e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 695/3000 1712/1712 [==============================] - 1s - loss: 8.4303e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 696/3000 1712/1712 [==============================] - 1s - loss: 8.8182e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 697/3000 1712/1712 [==============================] - 1s - loss: 7.9770e-04 - acc: 0.8254 - val_loss: 0.0017 - val_acc: 0.7921 Epoch 698/3000 1712/1712 [==============================] - 1s - loss: 8.3696e-04 - acc: 0.8283 - val_loss: 0.0015 - val_acc: 0.7944 Epoch 699/3000 1712/1712 [==============================] - 1s - loss: 8.7495e-04 - acc: 0.8394 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 700/3000 1712/1712 [==============================] - 1s - loss: 8.3687e-04 - acc: 0.8236 - val_loss: 0.0018 - val_acc: 0.7664 Epoch 701/3000 1712/1712 [==============================] - 1s - loss: 8.7138e-04 - acc: 0.8324 - val_loss: 0.0018 - val_acc: 0.7664 Epoch 702/3000 1712/1712 [==============================] - 1s - loss: 8.7947e-04 - acc: 0.8329 - val_loss: 0.0019 - val_acc: 0.7453 Epoch 703/3000 1712/1712 [==============================] - 1s - loss: 7.9417e-04 - acc: 0.8265 - val_loss: 0.0020 - val_acc: 0.7360 Epoch 704/3000 1712/1712 [==============================] - 1s - loss: 8.7479e-04 - acc: 0.8236 - val_loss: 0.0018 - val_acc: 0.7874 Epoch 705/3000 1712/1712 [==============================] - 1s - loss: 8.0251e-04 - acc: 0.8341 - val_loss: 0.0021 - val_acc: 0.7126 Epoch 706/3000 1712/1712 [==============================] - 1s - loss: 9.4000e-04 - acc: 0.8090 - val_loss: 0.0015 - val_acc: 0.7500 Epoch 707/3000 1712/1712 [==============================] - 1s - loss: 7.8316e-04 - acc: 0.8207 - val_loss: 0.0015 - val_acc: 0.7453 Epoch 708/3000 1712/1712 [==============================] - 1s - loss: 7.9957e-04 - acc: 0.8224 - val_loss: 0.0021 - val_acc: 0.7150 Epoch 709/3000 1712/1712 [==============================] - 1s - loss: 8.5172e-04 - acc: 0.8125 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 710/3000 1712/1712 [==============================] - 1s - loss: 8.7407e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7640 Epoch 711/3000 1712/1712 [==============================] - 1s - loss: 7.7919e-04 - acc: 0.8411 - val_loss: 0.0019 - val_acc: 0.7593 Epoch 712/3000 1712/1712 [==============================] - 1s - loss: 8.8932e-04 - acc: 0.8329 - val_loss: 0.0017 - val_acc: 0.8037 Epoch 713/3000 1712/1712 [==============================] - 1s - loss: 8.6057e-04 - acc: 0.8137 - val_loss: 0.0015 - val_acc: 0.7804 Epoch 714/3000 1712/1712 [==============================] - 1s - loss: 8.8247e-04 - acc: 0.8207 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 715/3000 1712/1712 [==============================] - 1s - loss: 9.2438e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 716/3000 1712/1712 [==============================] - 1s - loss: 8.0118e-04 - acc: 0.8394 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 717/3000 1712/1712 [==============================] - 1s - loss: 8.3462e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 718/3000 1712/1712 [==============================] - 1s - loss: 8.5577e-04 - acc: 0.8259 - val_loss: 0.0015 - val_acc: 0.7593 Epoch 719/3000 1712/1712 [==============================] - 1s - loss: 8.5221e-04 - acc: 0.8224 - val_loss: 0.0020 - val_acc: 0.7243 Epoch 720/3000 1712/1712 [==============================] - 1s - loss: 8.2249e-04 - acc: 0.8242 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 721/3000 1712/1712 [==============================] - 1s - loss: 8.6172e-04 - acc: 0.8411 - val_loss: 0.0016 - val_acc: 0.7734 Epoch 722/3000 1712/1712 [==============================] - 1s - loss: 8.4191e-04 - acc: 0.8283 - val_loss: 0.0019 - val_acc: 0.7266 Epoch 723/3000 1712/1712 [==============================] - 1s - loss: 9.1698e-04 - acc: 0.8370 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 724/3000 1712/1712 [==============================] - 1s - loss: 8.2538e-04 - acc: 0.8242 - val_loss: 0.0011 - val_acc: 0.8061 Epoch 725/3000 1712/1712 [==============================] - 1s - loss: 8.3629e-04 - acc: 0.8254 - val_loss: 0.0020 - val_acc: 0.7266 Epoch 726/3000 1712/1712 [==============================] - 1s - loss: 7.9948e-04 - acc: 0.8300 - val_loss: 0.0014 - val_acc: 0.8061 Epoch 727/3000 1712/1712 [==============================] - 1s - loss: 9.2709e-04 - acc: 0.8201 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 728/3000 1712/1712 [==============================] - 1s - loss: 8.6683e-04 - acc: 0.8113 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 729/3000 1712/1712 [==============================] - 1s - loss: 7.8897e-04 - acc: 0.8364 - val_loss: 0.0021 - val_acc: 0.7150 Epoch 730/3000 1712/1712 [==============================] - 1s - loss: 8.6918e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 731/3000 1712/1712 [==============================] - 1s - loss: 8.2308e-04 - acc: 0.8277 - val_loss: 0.0015 - val_acc: 0.8061 Epoch 732/3000 1712/1712 [==============================] - 1s - loss: 9.4655e-04 - acc: 0.8189 - val_loss: 0.0017 - val_acc: 0.7430 Epoch 733/3000 1712/1712 [==============================] - 1s - loss: 8.4514e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.8154 Epoch 734/3000 1712/1712 [==============================] - 1s - loss: 8.6888e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 735/3000 1712/1712 [==============================] - 1s - loss: 8.6704e-04 - acc: 0.8236 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 736/3000 1712/1712 [==============================] - 1s - loss: 8.6123e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 737/3000 1712/1712 [==============================] - 1s - loss: 9.0394e-04 - acc: 0.8207 - val_loss: 0.0014 - val_acc: 0.7547 Epoch 738/3000 1712/1712 [==============================] - 1s - loss: 7.9158e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 739/3000 1712/1712 [==============================] - 1s - loss: 8.1538e-04 - acc: 0.8312 - val_loss: 0.0020 - val_acc: 0.7617 Epoch 740/3000 1712/1712 [==============================] - 1s - loss: 8.4203e-04 - acc: 0.8178 - val_loss: 0.0020 - val_acc: 0.7360 Epoch 741/3000 1712/1712 [==============================] - 1s - loss: 8.1747e-04 - acc: 0.8213 - val_loss: 0.0015 - val_acc: 0.7710 Epoch 742/3000 1712/1712 [==============================] - 1s - loss: 8.0931e-04 - acc: 0.8300 - val_loss: 0.0020 - val_acc: 0.7243 Epoch 743/3000 1712/1712 [==============================] - 1s - loss: 9.1477e-04 - acc: 0.8329 - val_loss: 0.0014 - val_acc: 0.8014 Epoch 744/3000 1712/1712 [==============================] - 1s - loss: 8.5840e-04 - acc: 0.8213 - val_loss: 0.0014 - val_acc: 0.7967 Epoch 745/3000 1712/1712 [==============================] - 1s - loss: 8.0564e-04 - acc: 0.8435 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 746/3000 1712/1712 [==============================] - 1s - loss: 9.1076e-04 - acc: 0.8119 - val_loss: 0.0014 - val_acc: 0.8154 Epoch 747/3000 1712/1712 [==============================] - 1s - loss: 7.4038e-04 - acc: 0.8359 - val_loss: 0.0011 - val_acc: 0.8061 Epoch 748/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8172 - val_loss: 0.0018 - val_acc: 0.7500 Epoch 749/3000 1712/1712 [==============================] - 1s - loss: 8.0065e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 750/3000 1712/1712 [==============================] - 1s - loss: 9.1600e-04 - acc: 0.8201 - val_loss: 0.0015 - val_acc: 0.7897 Epoch 751/3000 1712/1712 [==============================] - 1s - loss: 8.2038e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 752/3000 1712/1712 [==============================] - 1s - loss: 8.3817e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.8178 Epoch 753/3000 1712/1712 [==============================] - 1s - loss: 8.3171e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 754/3000 1712/1712 [==============================] - 1s - loss: 8.9996e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 755/3000 1712/1712 [==============================] - 1s - loss: 7.9991e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 756/3000 1712/1712 [==============================] - 1s - loss: 8.8981e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 757/3000 1712/1712 [==============================] - 1s - loss: 8.4441e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.8107 Epoch 758/3000 1712/1712 [==============================] - 1s - loss: 7.7372e-04 - acc: 0.8388 - val_loss: 0.0017 - val_acc: 0.7850 Epoch 759/3000 1712/1712 [==============================] - 1s - loss: 9.8599e-04 - acc: 0.8160 - val_loss: 0.0016 - val_acc: 0.8037 Epoch 760/3000 1712/1712 [==============================] - 1s - loss: 8.1205e-04 - acc: 0.8312 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 761/3000 1712/1712 [==============================] - 1s - loss: 8.0580e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 762/3000 1712/1712 [==============================] - 1s - loss: 8.6703e-04 - acc: 0.8259 - val_loss: 0.0014 - val_acc: 0.7944 Epoch 763/3000 1712/1712 [==============================] - 1s - loss: 8.6957e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 764/3000 1712/1712 [==============================] - 1s - loss: 8.2992e-04 - acc: 0.8172 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 765/3000 1712/1712 [==============================] - 1s - loss: 8.3037e-04 - acc: 0.8341 - val_loss: 0.0018 - val_acc: 0.7500 Epoch 766/3000 1712/1712 [==============================] - 1s - loss: 8.6687e-04 - acc: 0.8271 - val_loss: 0.0020 - val_acc: 0.7430 Epoch 767/3000 1712/1712 [==============================] - 1s - loss: 8.0397e-04 - acc: 0.8224 - val_loss: 0.0023 - val_acc: 0.7150 Epoch 768/3000 1712/1712 [==============================] - 1s - loss: 8.1646e-04 - acc: 0.8329 - val_loss: 0.0011 - val_acc: 0.8084 Epoch 769/3000 1712/1712 [==============================] - 1s - loss: 8.5455e-04 - acc: 0.8324 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 770/3000 1712/1712 [==============================] - 1s - loss: 8.7076e-04 - acc: 0.8183 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 771/3000 1712/1712 [==============================] - 1s - loss: 8.4740e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 772/3000 1712/1712 [==============================] - 1s - loss: 8.3318e-04 - acc: 0.8271 - val_loss: 0.0011 - val_acc: 0.8154 Epoch 773/3000 1712/1712 [==============================] - 1s - loss: 8.2366e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 774/3000 1712/1712 [==============================] - 1s - loss: 9.0606e-04 - acc: 0.8353 - val_loss: 0.0011 - val_acc: 0.7991 Epoch 775/3000 1712/1712 [==============================] - 1s - loss: 7.9791e-04 - acc: 0.8324 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 776/3000 1712/1712 [==============================] - 1s - loss: 8.4214e-04 - acc: 0.8341 - val_loss: 0.0014 - val_acc: 0.7523 Epoch 777/3000 1712/1712 [==============================] - 1s - loss: 9.1698e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.8201 Epoch 778/3000 1712/1712 [==============================] - 1s - loss: 7.9688e-04 - acc: 0.8201 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 779/3000 1712/1712 [==============================] - 1s - loss: 8.4603e-04 - acc: 0.8294 - val_loss: 0.0019 - val_acc: 0.7874 Epoch 780/3000 1712/1712 [==============================] - 1s - loss: 7.9898e-04 - acc: 0.8254 - val_loss: 0.0022 - val_acc: 0.7500 Epoch 781/3000 1712/1712 [==============================] - 1s - loss: 8.9885e-04 - acc: 0.8248 - val_loss: 0.0021 - val_acc: 0.7243 Epoch 782/3000 1712/1712 [==============================] - 1s - loss: 7.9937e-04 - acc: 0.8201 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 783/3000 1712/1712 [==============================] - 1s - loss: 8.4696e-04 - acc: 0.8394 - val_loss: 0.0012 - val_acc: 0.7664 Epoch 784/3000 1712/1712 [==============================] - 1s - loss: 8.9126e-04 - acc: 0.8347 - val_loss: 0.0017 - val_acc: 0.8084 Epoch 785/3000 1712/1712 [==============================] - 1s - loss: 7.7671e-04 - acc: 0.8259 - val_loss: 0.0020 - val_acc: 0.7079 Epoch 786/3000 1712/1712 [==============================] - 1s - loss: 9.7359e-04 - acc: 0.8271 - val_loss: 0.0019 - val_acc: 0.7196 Epoch 787/3000 1712/1712 [==============================] - 1s - loss: 8.0956e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 788/3000 1712/1712 [==============================] - 1s - loss: 7.6774e-04 - acc: 0.8405 - val_loss: 0.0016 - val_acc: 0.7407 Epoch 789/3000 1712/1712 [==============================] - 1s - loss: 8.6953e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 790/3000 1712/1712 [==============================] - 1s - loss: 8.5868e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.8131 Epoch 791/3000 1712/1712 [==============================] - 1s - loss: 8.5456e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 792/3000 1712/1712 [==============================] - 1s - loss: 8.2969e-04 - acc: 0.8440 - val_loss: 0.0018 - val_acc: 0.7547 Epoch 793/3000 1712/1712 [==============================] - 1s - loss: 8.7240e-04 - acc: 0.8277 - val_loss: 0.0019 - val_acc: 0.7757 Epoch 794/3000 1712/1712 [==============================] - 1s - loss: 8.3780e-04 - acc: 0.8248 - val_loss: 0.0015 - val_acc: 0.7710 Epoch 795/3000 1712/1712 [==============================] - 1s - loss: 7.2229e-04 - acc: 0.8429 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 796/3000 1712/1712 [==============================] - 1s - loss: 0.0010 - acc: 0.8242 - val_loss: 0.0016 - val_acc: 0.7500 Epoch 797/3000 1712/1712 [==============================] - 1s - loss: 7.1144e-04 - acc: 0.8411 - val_loss: 0.0024 - val_acc: 0.7266 Epoch 798/3000 1712/1712 [==============================] - 1s - loss: 8.4462e-04 - acc: 0.8289 - val_loss: 0.0018 - val_acc: 0.7243 Epoch 799/3000 1712/1712 [==============================] - 1s - loss: 9.0108e-04 - acc: 0.8306 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 800/3000 1712/1712 [==============================] - 1s - loss: 8.6002e-04 - acc: 0.8113 - val_loss: 0.0018 - val_acc: 0.7593 Epoch 801/3000 1712/1712 [==============================] - 1s - loss: 8.1169e-04 - acc: 0.8359 - val_loss: 0.0021 - val_acc: 0.7664 Epoch 802/3000 1712/1712 [==============================] - 1s - loss: 8.6038e-04 - acc: 0.8376 - val_loss: 0.0023 - val_acc: 0.7383 Epoch 803/3000 1712/1712 [==============================] - 1s - loss: 8.4694e-04 - acc: 0.8329 - val_loss: 0.0019 - val_acc: 0.7664 Epoch 804/3000 1712/1712 [==============================] - 1s - loss: 8.1055e-04 - acc: 0.8230 - val_loss: 0.0021 - val_acc: 0.7313 Epoch 805/3000 1712/1712 [==============================] - 1s - loss: 8.6652e-04 - acc: 0.8335 - val_loss: 0.0018 - val_acc: 0.7500 Epoch 806/3000 1712/1712 [==============================] - 1s - loss: 9.2681e-04 - acc: 0.8242 - val_loss: 0.0018 - val_acc: 0.7103 Epoch 807/3000 1712/1712 [==============================] - 1s - loss: 8.0647e-04 - acc: 0.8207 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 808/3000 1712/1712 [==============================] - 1s - loss: 8.7097e-04 - acc: 0.8265 - val_loss: 0.0019 - val_acc: 0.7383 Epoch 809/3000 1712/1712 [==============================] - 1s - loss: 8.1207e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 810/3000 1712/1712 [==============================] - 1s - loss: 8.2847e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 811/3000 1712/1712 [==============================] - 1s - loss: 8.9188e-04 - acc: 0.8230 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 812/3000 1712/1712 [==============================] - 1s - loss: 8.7096e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 813/3000 1712/1712 [==============================] - 1s - loss: 8.2237e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 814/3000 1712/1712 [==============================] - 1s - loss: 8.5997e-04 - acc: 0.8306 - val_loss: 0.0014 - val_acc: 0.7617 Epoch 815/3000 1712/1712 [==============================] - 1s - loss: 8.5348e-04 - acc: 0.8300 - val_loss: 0.0020 - val_acc: 0.7266 Epoch 816/3000 1712/1712 [==============================] - 1s - loss: 8.5782e-04 - acc: 0.8254 - val_loss: 0.0016 - val_acc: 0.8014 Epoch 817/3000 1712/1712 [==============================] - 1s - loss: 8.6104e-04 - acc: 0.8242 - val_loss: 0.0015 - val_acc: 0.7827 Epoch 818/3000 1712/1712 [==============================] - 1s - loss: 7.5545e-04 - acc: 0.8265 - val_loss: 0.0023 - val_acc: 0.7173 Epoch 819/3000 1712/1712 [==============================] - 1s - loss: 8.9464e-04 - acc: 0.8312 - val_loss: 0.0019 - val_acc: 0.7640 Epoch 820/3000 1712/1712 [==============================] - 1s - loss: 8.9133e-04 - acc: 0.8294 - val_loss: 0.0020 - val_acc: 0.7173 Epoch 821/3000 1712/1712 [==============================] - 1s - loss: 8.6531e-04 - acc: 0.8400 - val_loss: 0.0018 - val_acc: 0.7383 Epoch 822/3000 1712/1712 [==============================] - 1s - loss: 8.6579e-04 - acc: 0.8300 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 823/3000 1712/1712 [==============================] - 1s - loss: 7.9203e-04 - acc: 0.8283 - val_loss: 0.0011 - val_acc: 0.8154 Epoch 824/3000 1712/1712 [==============================] - 1s - loss: 8.4275e-04 - acc: 0.8236 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 825/3000 1712/1712 [==============================] - 1s - loss: 8.6459e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 826/3000 1712/1712 [==============================] - 1s - loss: 8.4142e-04 - acc: 0.8405 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 827/3000 1712/1712 [==============================] - 1s - loss: 7.6506e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 828/3000 1712/1712 [==============================] - 1s - loss: 8.5536e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 829/3000 1712/1712 [==============================] - 1s - loss: 8.9494e-04 - acc: 0.8195 - val_loss: 0.0014 - val_acc: 0.8084 Epoch 830/3000 1712/1712 [==============================] - 1s - loss: 7.6513e-04 - acc: 0.8324 - val_loss: 0.0026 - val_acc: 0.7103 Epoch 831/3000 1712/1712 [==============================] - 1s - loss: 8.8433e-04 - acc: 0.8213 - val_loss: 0.0017 - val_acc: 0.7710 Epoch 832/3000 1712/1712 [==============================] - 1s - loss: 7.7277e-04 - acc: 0.8289 - val_loss: 0.0015 - val_acc: 0.7383 Epoch 833/3000 1712/1712 [==============================] - 1s - loss: 8.9024e-04 - acc: 0.8131 - val_loss: 0.0017 - val_acc: 0.7617 Epoch 834/3000 1712/1712 [==============================] - 1s - loss: 8.1984e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 835/3000 1712/1712 [==============================] - 1s - loss: 8.1045e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.8178 Epoch 836/3000 1712/1712 [==============================] - 1s - loss: 9.0685e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 837/3000 1712/1712 [==============================] - 1s - loss: 7.1998e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 838/3000 1712/1712 [==============================] - 1s - loss: 9.0441e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 839/3000 1712/1712 [==============================] - 1s - loss: 8.3359e-04 - acc: 0.8294 - val_loss: 0.0014 - val_acc: 0.7593 Epoch 840/3000 1712/1712 [==============================] - 1s - loss: 8.3037e-04 - acc: 0.8242 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 841/3000 1712/1712 [==============================] - 1s - loss: 8.5569e-04 - acc: 0.8353 - val_loss: 0.0020 - val_acc: 0.7103 Epoch 842/3000 1712/1712 [==============================] - 1s - loss: 8.7094e-04 - acc: 0.8277 - val_loss: 0.0017 - val_acc: 0.8014 Epoch 843/3000 1712/1712 [==============================] - 1s - loss: 8.1320e-04 - acc: 0.8300 - val_loss: 0.0019 - val_acc: 0.7290 Epoch 844/3000 1712/1712 [==============================] - 1s - loss: 8.2754e-04 - acc: 0.8324 - val_loss: 0.0016 - val_acc: 0.7383 Epoch 845/3000 1712/1712 [==============================] - 1s - loss: 8.7548e-04 - acc: 0.8143 - val_loss: 0.0020 - val_acc: 0.7336 Epoch 846/3000 1712/1712 [==============================] - 1s - loss: 8.5297e-04 - acc: 0.8283 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 847/3000 1712/1712 [==============================] - 1s - loss: 8.3951e-04 - acc: 0.8364 - val_loss: 0.0013 - val_acc: 0.8178 Epoch 848/3000 1712/1712 [==============================] - 1s - loss: 7.9134e-04 - acc: 0.8464 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 849/3000 1712/1712 [==============================] - 1s - loss: 9.6663e-04 - acc: 0.8259 - val_loss: 0.0015 - val_acc: 0.7523 Epoch 850/3000 1712/1712 [==============================] - 1s - loss: 7.9849e-04 - acc: 0.8400 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 851/3000 1712/1712 [==============================] - 1s - loss: 8.6812e-04 - acc: 0.8347 - val_loss: 0.0023 - val_acc: 0.7173 Epoch 852/3000 1712/1712 [==============================] - 1s - loss: 8.3356e-04 - acc: 0.8306 - val_loss: 0.0018 - val_acc: 0.7640 Epoch 853/3000 1712/1712 [==============================] - 1s - loss: 8.4099e-04 - acc: 0.8318 - val_loss: 0.0024 - val_acc: 0.7640 Epoch 854/3000 1712/1712 [==============================] - 1s - loss: 8.2836e-04 - acc: 0.8242 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 855/3000 1712/1712 [==============================] - 1s - loss: 8.7842e-04 - acc: 0.8277 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 856/3000 1712/1712 [==============================] - 1s - loss: 8.1180e-04 - acc: 0.8201 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 857/3000 1712/1712 [==============================] - 1s - loss: 8.8705e-04 - acc: 0.8143 - val_loss: 0.0024 - val_acc: 0.7523 Epoch 858/3000 1712/1712 [==============================] - 1s - loss: 8.5932e-04 - acc: 0.8411 - val_loss: 0.0018 - val_acc: 0.7500 Epoch 859/3000 1712/1712 [==============================] - 1s - loss: 7.5580e-04 - acc: 0.8370 - val_loss: 0.0019 - val_acc: 0.7500 Epoch 860/3000 1712/1712 [==============================] - 1s - loss: 9.8896e-04 - acc: 0.8201 - val_loss: 0.0016 - val_acc: 0.7874 Epoch 861/3000 1712/1712 [==============================] - 1s - loss: 8.0033e-04 - acc: 0.8324 - val_loss: 0.0021 - val_acc: 0.7103 Epoch 862/3000 1712/1712 [==============================] - 1s - loss: 9.1779e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 863/3000 1712/1712 [==============================] - 1s - loss: 8.2233e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 864/3000 1712/1712 [==============================] - 1s - loss: 8.5279e-04 - acc: 0.8335 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 865/3000 1712/1712 [==============================] - 1s - loss: 8.8853e-04 - acc: 0.8119 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 866/3000 1712/1712 [==============================] - 1s - loss: 7.6421e-04 - acc: 0.8306 - val_loss: 0.0015 - val_acc: 0.7500 Epoch 867/3000 1712/1712 [==============================] - 1s - loss: 8.7974e-04 - acc: 0.8107 - val_loss: 0.0018 - val_acc: 0.7850 Epoch 868/3000 1712/1712 [==============================] - 1s - loss: 8.1130e-04 - acc: 0.8294 - val_loss: 0.0018 - val_acc: 0.7757 Epoch 869/3000 1712/1712 [==============================] - 1s - loss: 7.8946e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 870/3000 1712/1712 [==============================] - 1s - loss: 8.5818e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.8154 Epoch 871/3000 1712/1712 [==============================] - 1s - loss: 9.3170e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.7664 Epoch 872/3000 1712/1712 [==============================] - 1s - loss: 7.8297e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.8131 Epoch 873/3000 1712/1712 [==============================] - 1s - loss: 8.4260e-04 - acc: 0.8481 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 874/3000 1712/1712 [==============================] - 1s - loss: 8.4673e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 875/3000 1712/1712 [==============================] - 1s - loss: 8.4084e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 876/3000 1712/1712 [==============================] - 1s - loss: 8.4368e-04 - acc: 0.8324 - val_loss: 0.0014 - val_acc: 0.7850 Epoch 877/3000 1712/1712 [==============================] - 1s - loss: 8.1978e-04 - acc: 0.8178 - val_loss: 0.0017 - val_acc: 0.7430 Epoch 878/3000 1712/1712 [==============================] - 1s - loss: 8.6891e-04 - acc: 0.8189 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 879/3000 1712/1712 [==============================] - 1s - loss: 8.1387e-04 - acc: 0.8230 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 880/3000 1712/1712 [==============================] - 1s - loss: 8.8135e-04 - acc: 0.8388 - val_loss: 0.0014 - val_acc: 0.7874 Epoch 881/3000 1712/1712 [==============================] - 1s - loss: 7.5461e-04 - acc: 0.8265 - val_loss: 0.0015 - val_acc: 0.7593 Epoch 882/3000 1712/1712 [==============================] - 1s - loss: 8.6001e-04 - acc: 0.8143 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 883/3000 1712/1712 [==============================] - 1s - loss: 8.7776e-04 - acc: 0.8213 - val_loss: 0.0024 - val_acc: 0.7033 Epoch 884/3000 1712/1712 [==============================] - 1s - loss: 8.7273e-04 - acc: 0.8271 - val_loss: 0.0015 - val_acc: 0.8037 Epoch 885/3000 1712/1712 [==============================] - 1s - loss: 8.2557e-04 - acc: 0.8417 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 886/3000 1712/1712 [==============================] - 1s - loss: 8.8183e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 887/3000 1712/1712 [==============================] - 1s - loss: 7.6861e-04 - acc: 0.8277 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 888/3000 1712/1712 [==============================] - 1s - loss: 8.9838e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.7664 Epoch 889/3000 1712/1712 [==============================] - 1s - loss: 8.6579e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 890/3000 1712/1712 [==============================] - 1s - loss: 7.7942e-04 - acc: 0.8271 - val_loss: 0.0020 - val_acc: 0.7360 Epoch 891/3000 1712/1712 [==============================] - 1s - loss: 8.5807e-04 - acc: 0.8376 - val_loss: 0.0018 - val_acc: 0.7523 Epoch 892/3000 1712/1712 [==============================] - 1s - loss: 7.6693e-04 - acc: 0.8341 - val_loss: 0.0016 - val_acc: 0.7500 Epoch 893/3000 1712/1712 [==============================] - 1s - loss: 9.0445e-04 - acc: 0.8400 - val_loss: 0.0018 - val_acc: 0.7944 Epoch 894/3000 1712/1712 [==============================] - 1s - loss: 8.5427e-04 - acc: 0.8265 - val_loss: 0.0016 - val_acc: 0.7570 Epoch 895/3000 1712/1712 [==============================] - 1s - loss: 8.4185e-04 - acc: 0.8283 - val_loss: 0.0018 - val_acc: 0.7383 Epoch 896/3000 1712/1712 [==============================] - 1s - loss: 8.5239e-04 - acc: 0.8294 - val_loss: 0.0018 - val_acc: 0.7523 Epoch 897/3000 1712/1712 [==============================] - 1s - loss: 8.2654e-04 - acc: 0.8254 - val_loss: 0.0017 - val_acc: 0.7897 Epoch 898/3000 1712/1712 [==============================] - 1s - loss: 7.9171e-04 - acc: 0.8411 - val_loss: 0.0025 - val_acc: 0.7243 Epoch 899/3000 1712/1712 [==============================] - 1s - loss: 8.8533e-04 - acc: 0.8259 - val_loss: 0.0020 - val_acc: 0.7477 Epoch 900/3000 1712/1712 [==============================] - 1s - loss: 8.0940e-04 - acc: 0.8242 - val_loss: 0.0014 - val_acc: 0.7570 Epoch 901/3000 1712/1712 [==============================] - 1s - loss: 8.3705e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 902/3000 1712/1712 [==============================] - 1s - loss: 8.3441e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.7664 Epoch 903/3000 1712/1712 [==============================] - 1s - loss: 8.8684e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 904/3000 1712/1712 [==============================] - 1s - loss: 7.9101e-04 - acc: 0.8335 - val_loss: 0.0019 - val_acc: 0.7243 Epoch 905/3000 1712/1712 [==============================] - 1s - loss: 8.7962e-04 - acc: 0.8172 - val_loss: 0.0018 - val_acc: 0.7290 Epoch 906/3000 1712/1712 [==============================] - 1s - loss: 8.7118e-04 - acc: 0.8224 - val_loss: 0.0015 - val_acc: 0.7734 Epoch 907/3000 1712/1712 [==============================] - 1s - loss: 8.2125e-04 - acc: 0.8283 - val_loss: 0.0014 - val_acc: 0.7593 Epoch 908/3000 1712/1712 [==============================] - 1s - loss: 7.9010e-04 - acc: 0.8230 - val_loss: 0.0018 - val_acc: 0.7500 Epoch 909/3000 1712/1712 [==============================] - 1s - loss: 8.1851e-04 - acc: 0.8259 - val_loss: 0.0020 - val_acc: 0.7126 Epoch 910/3000 1712/1712 [==============================] - 1s - loss: 8.3556e-04 - acc: 0.8218 - val_loss: 0.0022 - val_acc: 0.7103 Epoch 911/3000 1712/1712 [==============================] - 1s - loss: 8.8733e-04 - acc: 0.8207 - val_loss: 0.0015 - val_acc: 0.7290 Epoch 912/3000 1712/1712 [==============================] - 1s - loss: 8.1012e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7593 Epoch 913/3000 1712/1712 [==============================] - 1s - loss: 8.5964e-04 - acc: 0.8394 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 914/3000 1712/1712 [==============================] - 1s - loss: 9.0300e-04 - acc: 0.8259 - val_loss: 0.0013 - val_acc: 0.7640 Epoch 915/3000 1712/1712 [==============================] - 1s - loss: 7.7971e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 916/3000 1712/1712 [==============================] - 1s - loss: 7.9487e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 917/3000 1712/1712 [==============================] - 1s - loss: 8.9386e-04 - acc: 0.8400 - val_loss: 0.0015 - val_acc: 0.7336 Epoch 918/3000 1712/1712 [==============================] - 1s - loss: 8.1155e-04 - acc: 0.8359 - val_loss: 0.0015 - val_acc: 0.7547 Epoch 919/3000 1712/1712 [==============================] - 1s - loss: 7.9298e-04 - acc: 0.8289 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 920/3000 1712/1712 [==============================] - 1s - loss: 8.0629e-04 - acc: 0.8254 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 921/3000 1712/1712 [==============================] - 1s - loss: 7.8751e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 922/3000 1712/1712 [==============================] - 1s - loss: 8.7463e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 923/3000 1712/1712 [==============================] - 1s - loss: 7.9317e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 924/3000 1712/1712 [==============================] - 1s - loss: 9.0521e-04 - acc: 0.8528 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 925/3000 1712/1712 [==============================] - 1s - loss: 7.7028e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 926/3000 1712/1712 [==============================] - 1s - loss: 8.0489e-04 - acc: 0.8400 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 927/3000 1712/1712 [==============================] - 1s - loss: 8.4682e-04 - acc: 0.8324 - val_loss: 0.0018 - val_acc: 0.7780 Epoch 928/3000 1712/1712 [==============================] - 1s - loss: 7.9701e-04 - acc: 0.8370 - val_loss: 0.0019 - val_acc: 0.7430 Epoch 929/3000 1712/1712 [==============================] - 1s - loss: 8.0026e-04 - acc: 0.8259 - val_loss: 0.0023 - val_acc: 0.7243 Epoch 930/3000 1712/1712 [==============================] - 1s - loss: 9.0466e-04 - acc: 0.8347 - val_loss: 0.0014 - val_acc: 0.7967 Epoch 931/3000 1712/1712 [==============================] - 1s - loss: 7.8754e-04 - acc: 0.8411 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 932/3000 1712/1712 [==============================] - 1s - loss: 8.3823e-04 - acc: 0.8370 - val_loss: 0.0018 - val_acc: 0.7360 Epoch 933/3000 1712/1712 [==============================] - 1s - loss: 8.3461e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 934/3000 1712/1712 [==============================] - 1s - loss: 8.6231e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 935/3000 1712/1712 [==============================] - 1s - loss: 8.2599e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 936/3000 1712/1712 [==============================] - 1s - loss: 8.4474e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 937/3000 1712/1712 [==============================] - 1s - loss: 8.4311e-04 - acc: 0.8248 - val_loss: 0.0018 - val_acc: 0.7430 Epoch 938/3000 1712/1712 [==============================] - 1s - loss: 7.8300e-04 - acc: 0.8364 - val_loss: 0.0016 - val_acc: 0.7687 Epoch 939/3000 1712/1712 [==============================] - 1s - loss: 8.6762e-04 - acc: 0.8359 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 940/3000 1712/1712 [==============================] - 1s - loss: 7.9579e-04 - acc: 0.8400 - val_loss: 0.0019 - val_acc: 0.7360 Epoch 941/3000 1712/1712 [==============================] - 1s - loss: 8.5639e-04 - acc: 0.8248 - val_loss: 0.0021 - val_acc: 0.7453 Epoch 942/3000 1712/1712 [==============================] - 1s - loss: 8.2562e-04 - acc: 0.8300 - val_loss: 0.0023 - val_acc: 0.7173 Epoch 943/3000 1712/1712 [==============================] - 1s - loss: 7.4898e-04 - acc: 0.8516 - val_loss: 0.0019 - val_acc: 0.7383 Epoch 944/3000 1712/1712 [==============================] - 1s - loss: 8.9320e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 945/3000 1712/1712 [==============================] - 1s - loss: 8.9080e-04 - acc: 0.8236 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 946/3000 1712/1712 [==============================] - 1s - loss: 8.9480e-04 - acc: 0.8207 - val_loss: 0.0014 - val_acc: 0.7827 Epoch 947/3000 1712/1712 [==============================] - 1s - loss: 7.9669e-04 - acc: 0.8347 - val_loss: 0.0016 - val_acc: 0.7804 Epoch 948/3000 1712/1712 [==============================] - 1s - loss: 9.1017e-04 - acc: 0.8259 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 949/3000 1712/1712 [==============================] - 1s - loss: 8.7815e-04 - acc: 0.8300 - val_loss: 0.0016 - val_acc: 0.7944 Epoch 950/3000 1712/1712 [==============================] - 1s - loss: 7.5805e-04 - acc: 0.8435 - val_loss: 0.0020 - val_acc: 0.7500 Epoch 951/3000 1712/1712 [==============================] - 1s - loss: 9.0119e-04 - acc: 0.8248 - val_loss: 0.0023 - val_acc: 0.7150 Epoch 952/3000 1712/1712 [==============================] - 1s - loss: 7.3032e-04 - acc: 0.8411 - val_loss: 0.0013 - val_acc: 0.7383 Epoch 953/3000 1712/1712 [==============================] - 1s - loss: 8.6085e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 954/3000 1712/1712 [==============================] - 1s - loss: 9.4800e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 955/3000 1712/1712 [==============================] - 1s - loss: 7.9051e-04 - acc: 0.8341 - val_loss: 0.0014 - val_acc: 0.7757 Epoch 956/3000 1712/1712 [==============================] - 1s - loss: 8.1266e-04 - acc: 0.8271 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 957/3000 1712/1712 [==============================] - 1s - loss: 8.1152e-04 - acc: 0.8242 - val_loss: 0.0026 - val_acc: 0.7150 Epoch 958/3000 1712/1712 [==============================] - 1s - loss: 8.4730e-04 - acc: 0.8306 - val_loss: 0.0020 - val_acc: 0.7126 Epoch 959/3000 1712/1712 [==============================] - 1s - loss: 8.7011e-04 - acc: 0.8166 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 960/3000 1712/1712 [==============================] - 1s - loss: 7.9704e-04 - acc: 0.8335 - val_loss: 0.0021 - val_acc: 0.7313 Epoch 961/3000 1712/1712 [==============================] - 1s - loss: 8.5966e-04 - acc: 0.8218 - val_loss: 0.0018 - val_acc: 0.7407 Epoch 962/3000 1712/1712 [==============================] - 1s - loss: 8.5142e-04 - acc: 0.8306 - val_loss: 0.0011 - val_acc: 0.7710 Epoch 963/3000 1712/1712 [==============================] - 1s - loss: 7.7736e-04 - acc: 0.8195 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 964/3000 1712/1712 [==============================] - 1s - loss: 8.4551e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 965/3000 1712/1712 [==============================] - 1s - loss: 9.3887e-04 - acc: 0.8201 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 966/3000 1712/1712 [==============================] - 1s - loss: 7.7419e-04 - acc: 0.8452 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 967/3000 1712/1712 [==============================] - 1s - loss: 8.7330e-04 - acc: 0.8090 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 968/3000 1712/1712 [==============================] - 1s - loss: 8.3239e-04 - acc: 0.8218 - val_loss: 0.0014 - val_acc: 0.7734 Epoch 969/3000 1712/1712 [==============================] - 1s - loss: 8.7923e-04 - acc: 0.8376 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 970/3000 1712/1712 [==============================] - 1s - loss: 8.5677e-04 - acc: 0.8213 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 971/3000 1712/1712 [==============================] - 1s - loss: 8.4559e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 972/3000 1712/1712 [==============================] - 1s - loss: 9.0245e-04 - acc: 0.8277 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 973/3000 1712/1712 [==============================] - 1s - loss: 8.0459e-04 - acc: 0.8259 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 974/3000 1712/1712 [==============================] - 1s - loss: 7.8962e-04 - acc: 0.8440 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 975/3000 1712/1712 [==============================] - 1s - loss: 8.5074e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 976/3000 1712/1712 [==============================] - 1s - loss: 7.6177e-04 - acc: 0.8411 - val_loss: 0.0011 - val_acc: 0.8154 Epoch 977/3000 1712/1712 [==============================] - 1s - loss: 8.6011e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 978/3000 1712/1712 [==============================] - 1s - loss: 8.5917e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 979/3000 1712/1712 [==============================] - 1s - loss: 7.8249e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 980/3000 1712/1712 [==============================] - 1s - loss: 8.1966e-04 - acc: 0.8218 - val_loss: 0.0012 - val_acc: 0.8294 Epoch 981/3000 1712/1712 [==============================] - 1s - loss: 8.0313e-04 - acc: 0.8452 - val_loss: 0.0011 - val_acc: 0.8084 Epoch 982/3000 1712/1712 [==============================] - 1s - loss: 8.8074e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.8224 Epoch 983/3000 1712/1712 [==============================] - 1s - loss: 7.9241e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 984/3000 1712/1712 [==============================] - 1s - loss: 8.1956e-04 - acc: 0.8370 - val_loss: 0.0023 - val_acc: 0.7407 Epoch 985/3000 1712/1712 [==============================] - 1s - loss: 8.7749e-04 - acc: 0.8160 - val_loss: 0.0017 - val_acc: 0.7640 Epoch 986/3000 1712/1712 [==============================] - 1s - loss: 8.0457e-04 - acc: 0.8376 - val_loss: 0.0017 - val_acc: 0.7290 Epoch 987/3000 1712/1712 [==============================] - 1s - loss: 8.0958e-04 - acc: 0.8254 - val_loss: 0.0019 - val_acc: 0.7547 Epoch 988/3000 1712/1712 [==============================] - 1s - loss: 9.1094e-04 - acc: 0.8259 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 989/3000 1712/1712 [==============================] - 1s - loss: 8.9284e-04 - acc: 0.8207 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 990/3000 1712/1712 [==============================] - 1s - loss: 7.8569e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 991/3000 1712/1712 [==============================] - 1s - loss: 8.8158e-04 - acc: 0.8224 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 992/3000 1712/1712 [==============================] - 1s - loss: 7.8175e-04 - acc: 0.8195 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 993/3000 1712/1712 [==============================] - 1s - loss: 8.5268e-04 - acc: 0.8201 - val_loss: 0.0018 - val_acc: 0.7827 Epoch 994/3000 1712/1712 [==============================] - 1s - loss: 8.2095e-04 - acc: 0.8423 - val_loss: 0.0016 - val_acc: 0.7944 Epoch 995/3000 1712/1712 [==============================] - 1s - loss: 8.8006e-04 - acc: 0.8213 - val_loss: 0.0016 - val_acc: 0.7897 Epoch 996/3000 1712/1712 [==============================] - 1s - loss: 8.0224e-04 - acc: 0.8318 - val_loss: 0.0011 - val_acc: 0.8084 Epoch 997/3000 1712/1712 [==============================] - 1s - loss: 8.3380e-04 - acc: 0.8306 - val_loss: 0.0016 - val_acc: 0.7944 Epoch 998/3000 1712/1712 [==============================] - 1s - loss: 7.7653e-04 - acc: 0.8312 - val_loss: 0.0017 - val_acc: 0.7710 Epoch 999/3000 1712/1712 [==============================] - 1s - loss: 9.4243e-04 - acc: 0.8289 - val_loss: 0.0017 - val_acc: 0.7313 Epoch 1000/3000 1712/1712 [==============================] - 1s - loss: 7.8329e-04 - acc: 0.8300 - val_loss: 0.0014 - val_acc: 0.7874 Epoch 1001/3000 1712/1712 [==============================] - 1s - loss: 8.9083e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 1002/3000 1712/1712 [==============================] - 1s - loss: 7.6214e-04 - acc: 0.8230 - val_loss: 0.0010 - val_acc: 0.8178 Epoch 1003/3000 1712/1712 [==============================] - 1s - loss: 7.8363e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 1004/3000 1712/1712 [==============================] - 1s - loss: 8.7313e-04 - acc: 0.8359 - val_loss: 0.0023 - val_acc: 0.7150 Epoch 1005/3000 1712/1712 [==============================] - 1s - loss: 8.0974e-04 - acc: 0.8341 - val_loss: 0.0016 - val_acc: 0.7523 Epoch 1006/3000 1712/1712 [==============================] - 1s - loss: 9.2954e-04 - acc: 0.8306 - val_loss: 0.0016 - val_acc: 0.7617 Epoch 1007/3000 1712/1712 [==============================] - 1s - loss: 8.4386e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1008/3000 1712/1712 [==============================] - 1s - loss: 7.5239e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 1009/3000 1712/1712 [==============================] - 1s - loss: 8.1518e-04 - acc: 0.8218 - val_loss: 0.0015 - val_acc: 0.7453 Epoch 1010/3000 1712/1712 [==============================] - 1s - loss: 8.2284e-04 - acc: 0.8143 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 1011/3000 1712/1712 [==============================] - 1s - loss: 8.0537e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1012/3000 1712/1712 [==============================] - 1s - loss: 9.0336e-04 - acc: 0.8271 - val_loss: 0.0014 - val_acc: 0.7991 Epoch 1013/3000 1712/1712 [==============================] - 1s - loss: 8.0828e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1014/3000 1712/1712 [==============================] - 1s - loss: 8.7027e-04 - acc: 0.8143 - val_loss: 0.0012 - val_acc: 0.8154 Epoch 1015/3000 1712/1712 [==============================] - 1s - loss: 8.3202e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1016/3000 1712/1712 [==============================] - 1s - loss: 8.3227e-04 - acc: 0.8400 - val_loss: 0.0024 - val_acc: 0.7313 Epoch 1017/3000 1712/1712 [==============================] - 1s - loss: 8.1499e-04 - acc: 0.8364 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 1018/3000 1712/1712 [==============================] - 1s - loss: 9.0033e-04 - acc: 0.8324 - val_loss: 0.0017 - val_acc: 0.7734 Epoch 1019/3000 1712/1712 [==============================] - 1s - loss: 7.9974e-04 - acc: 0.8259 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1020/3000 1712/1712 [==============================] - 1s - loss: 8.6770e-04 - acc: 0.8353 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 1021/3000 1712/1712 [==============================] - 1s - loss: 8.5936e-04 - acc: 0.8294 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 1022/3000 1712/1712 [==============================] - 1s - loss: 7.7758e-04 - acc: 0.8417 - val_loss: 0.0017 - val_acc: 0.7874 Epoch 1023/3000 1712/1712 [==============================] - 1s - loss: 8.3654e-04 - acc: 0.8300 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 1024/3000 1712/1712 [==============================] - 1s - loss: 8.4279e-04 - acc: 0.8277 - val_loss: 0.0015 - val_acc: 0.7593 Epoch 1025/3000 1712/1712 [==============================] - 1s - loss: 8.7040e-04 - acc: 0.8312 - val_loss: 0.0016 - val_acc: 0.7360 Epoch 1026/3000 1712/1712 [==============================] - 1s - loss: 7.8972e-04 - acc: 0.8294 - val_loss: 0.0016 - val_acc: 0.7780 Epoch 1027/3000 1712/1712 [==============================] - 1s - loss: 8.5767e-04 - acc: 0.8283 - val_loss: 0.0015 - val_acc: 0.7710 Epoch 1028/3000 1712/1712 [==============================] - 1s - loss: 8.0092e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 1029/3000 1712/1712 [==============================] - 1s - loss: 8.9486e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1030/3000 1712/1712 [==============================] - 1s - loss: 8.2704e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 1031/3000 1712/1712 [==============================] - 1s - loss: 8.0754e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 1032/3000 1712/1712 [==============================] - 1s - loss: 7.6978e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 1033/3000 1712/1712 [==============================] - 1s - loss: 8.8372e-04 - acc: 0.8271 - val_loss: 0.0014 - val_acc: 0.7664 Epoch 1034/3000 1712/1712 [==============================] - 1s - loss: 8.0833e-04 - acc: 0.8189 - val_loss: 0.0017 - val_acc: 0.7360 Epoch 1035/3000 1712/1712 [==============================] - 1s - loss: 8.7287e-04 - acc: 0.8294 - val_loss: 0.0018 - val_acc: 0.7430 Epoch 1036/3000 1712/1712 [==============================] - 1s - loss: 8.0913e-04 - acc: 0.8189 - val_loss: 0.0020 - val_acc: 0.7570 Epoch 1037/3000 1712/1712 [==============================] - 1s - loss: 8.2794e-04 - acc: 0.8254 - val_loss: 0.0016 - val_acc: 0.7593 Epoch 1038/3000 1712/1712 [==============================] - 1s - loss: 8.2298e-04 - acc: 0.8113 - val_loss: 0.0016 - val_acc: 0.7547 Epoch 1039/3000 1712/1712 [==============================] - 1s - loss: 8.3673e-04 - acc: 0.8446 - val_loss: 0.0016 - val_acc: 0.7617 Epoch 1040/3000 1712/1712 [==============================] - 1s - loss: 8.2910e-04 - acc: 0.8166 - val_loss: 0.0017 - val_acc: 0.7780 Epoch 1041/3000 1712/1712 [==============================] - 1s - loss: 7.8396e-04 - acc: 0.8394 - val_loss: 0.0019 - val_acc: 0.7220 Epoch 1042/3000 1712/1712 [==============================] - 1s - loss: 8.1320e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 1043/3000 1712/1712 [==============================] - 1s - loss: 9.2551e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1044/3000 1712/1712 [==============================] - 1s - loss: 7.4085e-04 - acc: 0.8359 - val_loss: 0.0016 - val_acc: 0.8084 Epoch 1045/3000 1712/1712 [==============================] - 1s - loss: 8.6327e-04 - acc: 0.8236 - val_loss: 0.0021 - val_acc: 0.7150 Epoch 1046/3000 1712/1712 [==============================] - 1s - loss: 8.2878e-04 - acc: 0.8207 - val_loss: 0.0019 - val_acc: 0.7290 Epoch 1047/3000 1712/1712 [==============================] - 1s - loss: 8.3464e-04 - acc: 0.8271 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 1048/3000 1712/1712 [==============================] - 1s - loss: 8.4002e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1049/3000 1712/1712 [==============================] - 1s - loss: 7.5063e-04 - acc: 0.8417 - val_loss: 0.0017 - val_acc: 0.7196 Epoch 1050/3000 1712/1712 [==============================] - 1s - loss: 8.3444e-04 - acc: 0.8166 - val_loss: 0.0020 - val_acc: 0.7477 Epoch 1051/3000 1712/1712 [==============================] - 1s - loss: 8.0069e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1052/3000 1712/1712 [==============================] - 1s - loss: 8.6551e-04 - acc: 0.8289 - val_loss: 0.0018 - val_acc: 0.7079 Epoch 1053/3000 1712/1712 [==============================] - 1s - loss: 8.7552e-04 - acc: 0.8213 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 1054/3000 1712/1712 [==============================] - 1s - loss: 8.3133e-04 - acc: 0.8306 - val_loss: 0.0017 - val_acc: 0.7360 Epoch 1055/3000 1712/1712 [==============================] - 1s - loss: 7.5868e-04 - acc: 0.8248 - val_loss: 0.0020 - val_acc: 0.7500 Epoch 1056/3000 1712/1712 [==============================] - 1s - loss: 8.8331e-04 - acc: 0.8283 - val_loss: 0.0018 - val_acc: 0.7126 Epoch 1057/3000 1712/1712 [==============================] - 1s - loss: 7.5404e-04 - acc: 0.8435 - val_loss: 0.0015 - val_acc: 0.7897 Epoch 1058/3000 1712/1712 [==============================] - 1s - loss: 8.6752e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 1059/3000 1712/1712 [==============================] - 1s - loss: 8.4168e-04 - acc: 0.8423 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 1060/3000 1712/1712 [==============================] - 1s - loss: 8.4630e-04 - acc: 0.8335 - val_loss: 0.0011 - val_acc: 0.8131 Epoch 1061/3000 1712/1712 [==============================] - 1s - loss: 8.2142e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1062/3000 1712/1712 [==============================] - 1s - loss: 7.8694e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 1063/3000 1712/1712 [==============================] - 1s - loss: 8.8247e-04 - acc: 0.8248 - val_loss: 0.0011 - val_acc: 0.8061 Epoch 1064/3000 1712/1712 [==============================] - 1s - loss: 7.8547e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 1065/3000 1712/1712 [==============================] - 1s - loss: 8.0664e-04 - acc: 0.8172 - val_loss: 0.0020 - val_acc: 0.7780 Epoch 1066/3000 1712/1712 [==============================] - 1s - loss: 8.2987e-04 - acc: 0.8283 - val_loss: 0.0020 - val_acc: 0.7664 Epoch 1067/3000 1712/1712 [==============================] - 1s - loss: 8.2105e-04 - acc: 0.8505 - val_loss: 0.0011 - val_acc: 0.7850 Epoch 1068/3000 1712/1712 [==============================] - 1s - loss: 8.6315e-04 - acc: 0.8218 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 1069/3000 1712/1712 [==============================] - 1s - loss: 7.7875e-04 - acc: 0.8394 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 1070/3000 1712/1712 [==============================] - 1s - loss: 8.7992e-04 - acc: 0.8440 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 1071/3000 1712/1712 [==============================] - 1s - loss: 8.1781e-04 - acc: 0.8435 - val_loss: 0.0014 - val_acc: 0.7827 Epoch 1072/3000 1712/1712 [==============================] - 1s - loss: 7.8593e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1073/3000 1712/1712 [==============================] - 1s - loss: 8.3395e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1074/3000 1712/1712 [==============================] - 1s - loss: 8.3233e-04 - acc: 0.8213 - val_loss: 0.0011 - val_acc: 0.7664 Epoch 1075/3000 1712/1712 [==============================] - 1s - loss: 7.6642e-04 - acc: 0.8178 - val_loss: 0.0014 - val_acc: 0.7570 Epoch 1076/3000 1712/1712 [==============================] - 1s - loss: 8.5973e-04 - acc: 0.8172 - val_loss: 0.0011 - val_acc: 0.7874 Epoch 1077/3000 1712/1712 [==============================] - 1s - loss: 7.8341e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 1078/3000 1712/1712 [==============================] - 1s - loss: 8.3758e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 1079/3000 1712/1712 [==============================] - 1s - loss: 9.1362e-04 - acc: 0.8329 - val_loss: 0.0018 - val_acc: 0.7430 Epoch 1080/3000 1712/1712 [==============================] - 1s - loss: 7.4517e-04 - acc: 0.8493 - val_loss: 0.0022 - val_acc: 0.7336 Epoch 1081/3000 1712/1712 [==============================] - 1s - loss: 7.6267e-04 - acc: 0.8470 - val_loss: 0.0025 - val_acc: 0.7079 Epoch 1082/3000 1712/1712 [==============================] - 1s - loss: 8.7654e-04 - acc: 0.8213 - val_loss: 0.0022 - val_acc: 0.7430 Epoch 1083/3000 1712/1712 [==============================] - 1s - loss: 8.4463e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 1084/3000 1712/1712 [==============================] - 1s - loss: 8.7853e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 1085/3000 1712/1712 [==============================] - 1s - loss: 7.8884e-04 - acc: 0.8417 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1086/3000 1712/1712 [==============================] - 1s - loss: 8.6952e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.8178 Epoch 1087/3000 1712/1712 [==============================] - 1s - loss: 8.5701e-04 - acc: 0.8300 - val_loss: 0.0015 - val_acc: 0.7617 Epoch 1088/3000 1712/1712 [==============================] - 1s - loss: 8.5031e-04 - acc: 0.8294 - val_loss: 0.0017 - val_acc: 0.7313 Epoch 1089/3000 1712/1712 [==============================] - 1s - loss: 8.5671e-04 - acc: 0.8324 - val_loss: 0.0022 - val_acc: 0.7336 Epoch 1090/3000 1712/1712 [==============================] - 1s - loss: 8.0099e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1091/3000 1712/1712 [==============================] - 1s - loss: 8.3167e-04 - acc: 0.8283 - val_loss: 0.0011 - val_acc: 0.7874 Epoch 1092/3000 1712/1712 [==============================] - 1s - loss: 7.9007e-04 - acc: 0.8306 - val_loss: 0.0016 - val_acc: 0.7453 Epoch 1093/3000 1712/1712 [==============================] - 1s - loss: 8.3459e-04 - acc: 0.8382 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 1094/3000 1712/1712 [==============================] - 1s - loss: 8.7620e-04 - acc: 0.8242 - val_loss: 0.0018 - val_acc: 0.7313 Epoch 1095/3000 1712/1712 [==============================] - 1s - loss: 8.4250e-04 - acc: 0.8341 - val_loss: 0.0015 - val_acc: 0.7804 Epoch 1096/3000 1712/1712 [==============================] - 1s - loss: 8.1183e-04 - acc: 0.8446 - val_loss: 0.0012 - val_acc: 0.8248 Epoch 1097/3000 1712/1712 [==============================] - 1s - loss: 7.8833e-04 - acc: 0.8405 - val_loss: 0.0015 - val_acc: 0.7360 Epoch 1098/3000 1712/1712 [==============================] - 1s - loss: 8.9438e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1099/3000 1712/1712 [==============================] - 1s - loss: 8.1198e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 1100/3000 1712/1712 [==============================] - 1s - loss: 8.6169e-04 - acc: 0.8294 - val_loss: 0.0020 - val_acc: 0.7173 Epoch 1101/3000 1712/1712 [==============================] - 1s - loss: 7.5575e-04 - acc: 0.8481 - val_loss: 0.0025 - val_acc: 0.7079 Epoch 1102/3000 1712/1712 [==============================] - 1s - loss: 8.6822e-04 - acc: 0.8254 - val_loss: 0.0018 - val_acc: 0.7547 Epoch 1103/3000 1712/1712 [==============================] - 1s - loss: 8.3297e-04 - acc: 0.8376 - val_loss: 0.0018 - val_acc: 0.7383 Epoch 1104/3000 1712/1712 [==============================] - 1s - loss: 8.2066e-04 - acc: 0.8283 - val_loss: 0.0018 - val_acc: 0.7430 Epoch 1105/3000 1712/1712 [==============================] - 1s - loss: 7.9624e-04 - acc: 0.8283 - val_loss: 0.0022 - val_acc: 0.7126 Epoch 1106/3000 1712/1712 [==============================] - 1s - loss: 8.6271e-04 - acc: 0.8248 - val_loss: 0.0018 - val_acc: 0.7266 Epoch 1107/3000 1712/1712 [==============================] - 1s - loss: 8.0074e-04 - acc: 0.8329 - val_loss: 0.0017 - val_acc: 0.8084 Epoch 1108/3000 1712/1712 [==============================] - 1s - loss: 8.0811e-04 - acc: 0.8189 - val_loss: 0.0012 - val_acc: 0.8178 Epoch 1109/3000 1712/1712 [==============================] - 1s - loss: 8.6933e-04 - acc: 0.8160 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 1110/3000 1712/1712 [==============================] - 1s - loss: 7.5382e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.8178 Epoch 1111/3000 1712/1712 [==============================] - 1s - loss: 8.5797e-04 - acc: 0.8376 - val_loss: 0.0018 - val_acc: 0.7383 Epoch 1112/3000 1712/1712 [==============================] - 1s - loss: 7.9588e-04 - acc: 0.8265 - val_loss: 0.0020 - val_acc: 0.7313 Epoch 1113/3000 1712/1712 [==============================] - 1s - loss: 8.5806e-04 - acc: 0.8324 - val_loss: 0.0017 - val_acc: 0.7383 Epoch 1114/3000 1712/1712 [==============================] - 1s - loss: 7.9028e-04 - acc: 0.8259 - val_loss: 0.0017 - val_acc: 0.7220 Epoch 1115/3000 1712/1712 [==============================] - 1s - loss: 8.8972e-04 - acc: 0.8213 - val_loss: 0.0017 - val_acc: 0.7547 Epoch 1116/3000 1712/1712 [==============================] - 1s - loss: 8.0066e-04 - acc: 0.8382 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1117/3000 1712/1712 [==============================] - 1s - loss: 8.4584e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 1118/3000 1712/1712 [==============================] - 1s - loss: 8.0108e-04 - acc: 0.8242 - val_loss: 0.0014 - val_acc: 0.7360 Epoch 1119/3000 1712/1712 [==============================] - 1s - loss: 7.2907e-04 - acc: 0.8324 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 1120/3000 1712/1712 [==============================] - 1s - loss: 8.9788e-04 - acc: 0.8236 - val_loss: 0.0011 - val_acc: 0.8131 Epoch 1121/3000 1712/1712 [==============================] - 1s - loss: 8.6410e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.8178 Epoch 1122/3000 1712/1712 [==============================] - 1s - loss: 8.3150e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1123/3000 1712/1712 [==============================] - 1s - loss: 7.5033e-04 - acc: 0.8382 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 1124/3000 1712/1712 [==============================] - 1s - loss: 8.2687e-04 - acc: 0.8289 - val_loss: 0.0016 - val_acc: 0.7664 Epoch 1125/3000 1712/1712 [==============================] - 1s - loss: 8.6723e-04 - acc: 0.8400 - val_loss: 0.0022 - val_acc: 0.7126 Epoch 1126/3000 1712/1712 [==============================] - 1s - loss: 8.1297e-04 - acc: 0.8213 - val_loss: 0.0018 - val_acc: 0.7430 Epoch 1127/3000 1712/1712 [==============================] - 1s - loss: 8.4786e-04 - acc: 0.8294 - val_loss: 0.0018 - val_acc: 0.7453 Epoch 1128/3000 1712/1712 [==============================] - 1s - loss: 8.4024e-04 - acc: 0.8277 - val_loss: 0.0018 - val_acc: 0.7734 Epoch 1129/3000 1712/1712 [==============================] - 1s - loss: 8.5344e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1130/3000 1712/1712 [==============================] - 1s - loss: 7.9943e-04 - acc: 0.8353 - val_loss: 0.0014 - val_acc: 0.7500 Epoch 1131/3000 1712/1712 [==============================] - 1s - loss: 7.8167e-04 - acc: 0.8218 - val_loss: 0.0020 - val_acc: 0.7126 Epoch 1132/3000 1712/1712 [==============================] - 1s - loss: 8.0387e-04 - acc: 0.8359 - val_loss: 0.0020 - val_acc: 0.7313 Epoch 1133/3000 1712/1712 [==============================] - 1s - loss: 8.5194e-04 - acc: 0.8236 - val_loss: 0.0017 - val_acc: 0.7407 Epoch 1134/3000 1712/1712 [==============================] - 1s - loss: 8.2942e-04 - acc: 0.8382 - val_loss: 0.0018 - val_acc: 0.7967 Epoch 1135/3000 1712/1712 [==============================] - 1s - loss: 7.9270e-04 - acc: 0.8394 - val_loss: 0.0015 - val_acc: 0.7967 Epoch 1136/3000 1712/1712 [==============================] - 1s - loss: 8.1086e-04 - acc: 0.8324 - val_loss: 0.0019 - val_acc: 0.7360 Epoch 1137/3000 1712/1712 [==============================] - 1s - loss: 8.2621e-04 - acc: 0.8376 - val_loss: 0.0015 - val_acc: 0.7664 Epoch 1138/3000 1712/1712 [==============================] - 1s - loss: 9.0151e-04 - acc: 0.8335 - val_loss: 0.0020 - val_acc: 0.7360 Epoch 1139/3000 1712/1712 [==============================] - 1s - loss: 7.7104e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 1140/3000 1712/1712 [==============================] - 1s - loss: 8.5980e-04 - acc: 0.8388 - val_loss: 0.0011 - val_acc: 0.7921 Epoch 1141/3000 1712/1712 [==============================] - 1s - loss: 8.6345e-04 - acc: 0.8283 - val_loss: 0.0016 - val_acc: 0.7220 Epoch 1142/3000 1712/1712 [==============================] - 1s - loss: 8.6533e-04 - acc: 0.8242 - val_loss: 0.0017 - val_acc: 0.7407 Epoch 1143/3000 1712/1712 [==============================] - 1s - loss: 7.5924e-04 - acc: 0.8411 - val_loss: 0.0013 - val_acc: 0.7430 Epoch 1144/3000 1712/1712 [==============================] - 1s - loss: 8.2896e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 1145/3000 1712/1712 [==============================] - 1s - loss: 8.1270e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1146/3000 1712/1712 [==============================] - 1s - loss: 8.5282e-04 - acc: 0.8230 - val_loss: 0.0015 - val_acc: 0.7453 Epoch 1147/3000 1712/1712 [==============================] - 1s - loss: 8.1362e-04 - acc: 0.8347 - val_loss: 0.0017 - val_acc: 0.7570 Epoch 1148/3000 1712/1712 [==============================] - 1s - loss: 8.7784e-04 - acc: 0.8271 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 1149/3000 1712/1712 [==============================] - 1s - loss: 8.0171e-04 - acc: 0.8370 - val_loss: 0.0019 - val_acc: 0.7126 Epoch 1150/3000 1712/1712 [==============================] - 1s - loss: 7.5342e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7640 Epoch 1151/3000 1712/1712 [==============================] - 1s - loss: 8.3557e-04 - acc: 0.8499 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1152/3000 1712/1712 [==============================] - 1s - loss: 8.2721e-04 - acc: 0.8300 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1153/3000 1712/1712 [==============================] - 1s - loss: 7.7053e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1154/3000 1712/1712 [==============================] - 1s - loss: 8.7390e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1155/3000 1712/1712 [==============================] - 1s - loss: 8.1853e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 1156/3000 1712/1712 [==============================] - 1s - loss: 8.1975e-04 - acc: 0.8376 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 1157/3000 1712/1712 [==============================] - 1s - loss: 7.8078e-04 - acc: 0.8440 - val_loss: 0.0012 - val_acc: 0.7593 Epoch 1158/3000 1712/1712 [==============================] - 1s - loss: 8.7519e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1159/3000 1712/1712 [==============================] - 1s - loss: 7.5181e-04 - acc: 0.8394 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1160/3000 1712/1712 [==============================] - 1s - loss: 8.4193e-04 - acc: 0.8236 - val_loss: 0.0011 - val_acc: 0.7687 Epoch 1161/3000 1712/1712 [==============================] - 1s - loss: 7.8089e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 1162/3000 1712/1712 [==============================] - 1s - loss: 8.0284e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1163/3000 1712/1712 [==============================] - 1s - loss: 8.9164e-04 - acc: 0.8061 - val_loss: 0.0020 - val_acc: 0.7313 Epoch 1164/3000 1712/1712 [==============================] - 1s - loss: 8.0715e-04 - acc: 0.8370 - val_loss: 0.0022 - val_acc: 0.7220 Epoch 1165/3000 1712/1712 [==============================] - 1s - loss: 8.2990e-04 - acc: 0.8411 - val_loss: 0.0016 - val_acc: 0.7477 Epoch 1166/3000 1712/1712 [==============================] - 1s - loss: 8.4333e-04 - acc: 0.8329 - val_loss: 0.0014 - val_acc: 0.7921 Epoch 1167/3000 1712/1712 [==============================] - 1s - loss: 8.3505e-04 - acc: 0.8230 - val_loss: 0.0014 - val_acc: 0.7640 Epoch 1168/3000 1712/1712 [==============================] - 1s - loss: 7.7275e-04 - acc: 0.8341 - val_loss: 0.0025 - val_acc: 0.7290 Epoch 1169/3000 1712/1712 [==============================] - 1s - loss: 8.4038e-04 - acc: 0.8306 - val_loss: 0.0026 - val_acc: 0.7009 Epoch 1170/3000 1712/1712 [==============================] - 1s - loss: 8.1815e-04 - acc: 0.8289 - val_loss: 0.0015 - val_acc: 0.7687 Epoch 1171/3000 1712/1712 [==============================] - 1s - loss: 8.1658e-04 - acc: 0.8429 - val_loss: 0.0018 - val_acc: 0.7407 Epoch 1172/3000 1712/1712 [==============================] - 1s - loss: 8.8737e-04 - acc: 0.8289 - val_loss: 0.0016 - val_acc: 0.7734 Epoch 1173/3000 1712/1712 [==============================] - 1s - loss: 8.1088e-04 - acc: 0.8411 - val_loss: 0.0016 - val_acc: 0.7757 Epoch 1174/3000 1712/1712 [==============================] - 1s - loss: 8.2011e-04 - acc: 0.8312 - val_loss: 0.0020 - val_acc: 0.7500 Epoch 1175/3000 1712/1712 [==============================] - 1s - loss: 8.3330e-04 - acc: 0.8283 - val_loss: 0.0017 - val_acc: 0.8037 Epoch 1176/3000 1712/1712 [==============================] - 1s - loss: 8.4845e-04 - acc: 0.8382 - val_loss: 0.0021 - val_acc: 0.7173 Epoch 1177/3000 1712/1712 [==============================] - 1s - loss: 8.4960e-04 - acc: 0.8300 - val_loss: 0.0023 - val_acc: 0.7033 Epoch 1178/3000 1712/1712 [==============================] - 1s - loss: 7.8614e-04 - acc: 0.8370 - val_loss: 0.0019 - val_acc: 0.7336 Epoch 1179/3000 1712/1712 [==============================] - 1s - loss: 8.0212e-04 - acc: 0.8364 - val_loss: 0.0021 - val_acc: 0.7313 Epoch 1180/3000 1712/1712 [==============================] - 1s - loss: 9.4628e-04 - acc: 0.8312 - val_loss: 0.0018 - val_acc: 0.7150 Epoch 1181/3000 1712/1712 [==============================] - 1s - loss: 7.9339e-04 - acc: 0.8364 - val_loss: 0.0018 - val_acc: 0.7360 Epoch 1182/3000 1712/1712 [==============================] - 1s - loss: 8.4414e-04 - acc: 0.8289 - val_loss: 0.0023 - val_acc: 0.7126 Epoch 1183/3000 1712/1712 [==============================] - 1s - loss: 7.9030e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1184/3000 1712/1712 [==============================] - 1s - loss: 8.8965e-04 - acc: 0.8218 - val_loss: 0.0018 - val_acc: 0.7593 Epoch 1185/3000 1712/1712 [==============================] - 1s - loss: 7.5733e-04 - acc: 0.8388 - val_loss: 0.0023 - val_acc: 0.7079 Epoch 1186/3000 1712/1712 [==============================] - 1s - loss: 8.4051e-04 - acc: 0.8213 - val_loss: 0.0012 - val_acc: 0.7617 Epoch 1187/3000 1712/1712 [==============================] - 1s - loss: 8.4267e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 1188/3000 1712/1712 [==============================] - 1s - loss: 8.7467e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 1189/3000 1712/1712 [==============================] - 1s - loss: 8.4192e-04 - acc: 0.8242 - val_loss: 0.0016 - val_acc: 0.7523 Epoch 1190/3000 1712/1712 [==============================] - 1s - loss: 7.7166e-04 - acc: 0.8271 - val_loss: 0.0019 - val_acc: 0.7547 Epoch 1191/3000 1712/1712 [==============================] - 1s - loss: 8.2668e-04 - acc: 0.8294 - val_loss: 0.0019 - val_acc: 0.7336 Epoch 1192/3000 1712/1712 [==============================] - 1s - loss: 8.4661e-04 - acc: 0.8242 - val_loss: 0.0011 - val_acc: 0.7874 Epoch 1193/3000 1712/1712 [==============================] - 1s - loss: 8.8793e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 1194/3000 1712/1712 [==============================] - 1s - loss: 7.7761e-04 - acc: 0.8516 - val_loss: 0.0012 - val_acc: 0.8294 Epoch 1195/3000 1712/1712 [==============================] - 1s - loss: 8.0404e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.8201 Epoch 1196/3000 1712/1712 [==============================] - 1s - loss: 8.6209e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 1197/3000 1712/1712 [==============================] - 1s - loss: 7.8423e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1198/3000 1712/1712 [==============================] - 1s - loss: 8.3308e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1199/3000 1712/1712 [==============================] - 1s - loss: 7.8387e-04 - acc: 0.8446 - val_loss: 0.0012 - val_acc: 0.7664 Epoch 1200/3000 1712/1712 [==============================] - 1s - loss: 8.6813e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 1201/3000 1712/1712 [==============================] - 1s - loss: 8.3773e-04 - acc: 0.8429 - val_loss: 0.0015 - val_acc: 0.7967 Epoch 1202/3000 1712/1712 [==============================] - 1s - loss: 8.6617e-04 - acc: 0.8195 - val_loss: 0.0014 - val_acc: 0.7991 Epoch 1203/3000 1712/1712 [==============================] - 1s - loss: 7.6319e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 1204/3000 1712/1712 [==============================] - 1s - loss: 8.4333e-04 - acc: 0.8300 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 1205/3000 1712/1712 [==============================] - 1s - loss: 9.0593e-04 - acc: 0.8137 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 1206/3000 1712/1712 [==============================] - 1s - loss: 7.9842e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 1207/3000 1712/1712 [==============================] - 1s - loss: 8.3956e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.8131 Epoch 1208/3000 1712/1712 [==============================] - 1s - loss: 8.0421e-04 - acc: 0.8300 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1209/3000 1712/1712 [==============================] - 1s - loss: 8.0201e-04 - acc: 0.8458 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 1210/3000 1712/1712 [==============================] - 1s - loss: 8.6625e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1211/3000 1712/1712 [==============================] - 1s - loss: 8.7294e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1212/3000 1712/1712 [==============================] - 1s - loss: 8.3757e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 1213/3000 1712/1712 [==============================] - 1s - loss: 7.4583e-04 - acc: 0.8382 - val_loss: 0.0021 - val_acc: 0.7079 Epoch 1214/3000 1712/1712 [==============================] - 1s - loss: 8.8224e-04 - acc: 0.8329 - val_loss: 0.0017 - val_acc: 0.7850 Epoch 1215/3000 1712/1712 [==============================] - 1s - loss: 8.4987e-04 - acc: 0.8364 - val_loss: 0.0021 - val_acc: 0.7126 Epoch 1216/3000 1712/1712 [==============================] - 1s - loss: 8.2183e-04 - acc: 0.8277 - val_loss: 0.0021 - val_acc: 0.7500 Epoch 1217/3000 1712/1712 [==============================] - 1s - loss: 8.1353e-04 - acc: 0.8329 - val_loss: 0.0016 - val_acc: 0.7407 Epoch 1218/3000 1712/1712 [==============================] - 1s - loss: 7.1490e-04 - acc: 0.8400 - val_loss: 0.0020 - val_acc: 0.7500 Epoch 1219/3000 1712/1712 [==============================] - 1s - loss: 9.2355e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 1220/3000 1712/1712 [==============================] - 1s - loss: 7.8062e-04 - acc: 0.8429 - val_loss: 0.0016 - val_acc: 0.7407 Epoch 1221/3000 1712/1712 [==============================] - 1s - loss: 8.7495e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1222/3000 1712/1712 [==============================] - 1s - loss: 8.3362e-04 - acc: 0.8242 - val_loss: 0.0017 - val_acc: 0.7407 Epoch 1223/3000 1712/1712 [==============================] - 1s - loss: 8.1005e-04 - acc: 0.8265 - val_loss: 0.0014 - val_acc: 0.7874 Epoch 1224/3000 1712/1712 [==============================] - 1s - loss: 8.0705e-04 - acc: 0.8329 - val_loss: 0.0014 - val_acc: 0.7664 Epoch 1225/3000 1712/1712 [==============================] - 1s - loss: 8.8539e-04 - acc: 0.8248 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 1226/3000 1712/1712 [==============================] - 1s - loss: 8.2553e-04 - acc: 0.8429 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 1227/3000 1712/1712 [==============================] - 1s - loss: 7.8667e-04 - acc: 0.8183 - val_loss: 0.0014 - val_acc: 0.7430 Epoch 1228/3000 1712/1712 [==============================] - 1s - loss: 7.7357e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1229/3000 1712/1712 [==============================] - 1s - loss: 8.9233e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 1230/3000 1712/1712 [==============================] - 1s - loss: 7.7683e-04 - acc: 0.8429 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 1231/3000 1712/1712 [==============================] - 1s - loss: 7.8323e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 1232/3000 1712/1712 [==============================] - 1s - loss: 8.2939e-04 - acc: 0.8440 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 1233/3000 1712/1712 [==============================] - 1s - loss: 8.0895e-04 - acc: 0.8546 - val_loss: 0.0016 - val_acc: 0.7383 Epoch 1234/3000 1712/1712 [==============================] - 1s - loss: 8.7772e-04 - acc: 0.8271 - val_loss: 0.0016 - val_acc: 0.7967 Epoch 1235/3000 1712/1712 [==============================] - 1s - loss: 7.7528e-04 - acc: 0.8394 - val_loss: 0.0015 - val_acc: 0.7430 Epoch 1236/3000 1712/1712 [==============================] - 1s - loss: 8.5609e-04 - acc: 0.8306 - val_loss: 0.0014 - val_acc: 0.7734 Epoch 1237/3000 1712/1712 [==============================] - 1s - loss: 7.7844e-04 - acc: 0.8259 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 1238/3000 1712/1712 [==============================] - 1s - loss: 8.4627e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 1239/3000 1712/1712 [==============================] - 1s - loss: 8.0439e-04 - acc: 0.8400 - val_loss: 0.0021 - val_acc: 0.7360 Epoch 1240/3000 1712/1712 [==============================] - 1s - loss: 8.4291e-04 - acc: 0.8283 - val_loss: 0.0016 - val_acc: 0.7593 Epoch 1241/3000 1712/1712 [==============================] - 1s - loss: 7.9899e-04 - acc: 0.8224 - val_loss: 0.0017 - val_acc: 0.7383 Epoch 1242/3000 1712/1712 [==============================] - 1s - loss: 8.7216e-04 - acc: 0.8195 - val_loss: 0.0016 - val_acc: 0.7313 Epoch 1243/3000 1712/1712 [==============================] - 1s - loss: 8.9870e-04 - acc: 0.8213 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 1244/3000 1712/1712 [==============================] - 1s - loss: 7.6297e-04 - acc: 0.8341 - val_loss: 0.0014 - val_acc: 0.7850 Epoch 1245/3000 1712/1712 [==============================] - 1s - loss: 8.8069e-04 - acc: 0.8359 - val_loss: 0.0016 - val_acc: 0.7827 Epoch 1246/3000 1712/1712 [==============================] - 1s - loss: 7.6788e-04 - acc: 0.8359 - val_loss: 0.0016 - val_acc: 0.7897 Epoch 1247/3000 1712/1712 [==============================] - 1s - loss: 8.9747e-04 - acc: 0.8283 - val_loss: 0.0021 - val_acc: 0.7103 Epoch 1248/3000 1712/1712 [==============================] - 1s - loss: 7.1559e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1249/3000 1712/1712 [==============================] - 1s - loss: 9.4255e-04 - acc: 0.8394 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1250/3000 1712/1712 [==============================] - 1s - loss: 7.6751e-04 - acc: 0.8347 - val_loss: 0.0021 - val_acc: 0.7313 Epoch 1251/3000 1712/1712 [==============================] - 1s - loss: 8.6408e-04 - acc: 0.8265 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 1252/3000 1712/1712 [==============================] - 1s - loss: 8.2446e-04 - acc: 0.8353 - val_loss: 0.0018 - val_acc: 0.7804 Epoch 1253/3000 1712/1712 [==============================] - 1s - loss: 8.0179e-04 - acc: 0.8359 - val_loss: 0.0018 - val_acc: 0.7523 Epoch 1254/3000 1712/1712 [==============================] - 1s - loss: 8.4545e-04 - acc: 0.8277 - val_loss: 0.0021 - val_acc: 0.7430 Epoch 1255/3000 1712/1712 [==============================] - 1s - loss: 7.9610e-04 - acc: 0.8376 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 1256/3000 1712/1712 [==============================] - 1s - loss: 8.2506e-04 - acc: 0.8376 - val_loss: 0.0021 - val_acc: 0.7056 Epoch 1257/3000 1712/1712 [==============================] - 1s - loss: 8.3252e-04 - acc: 0.8435 - val_loss: 0.0017 - val_acc: 0.7804 Epoch 1258/3000 1712/1712 [==============================] - 1s - loss: 8.3789e-04 - acc: 0.8318 - val_loss: 0.0015 - val_acc: 0.7430 Epoch 1259/3000 1712/1712 [==============================] - 1s - loss: 7.8815e-04 - acc: 0.8400 - val_loss: 0.0019 - val_acc: 0.7313 Epoch 1260/3000 1712/1712 [==============================] - 1s - loss: 8.7650e-04 - acc: 0.8341 - val_loss: 0.0014 - val_acc: 0.7500 Epoch 1261/3000 1712/1712 [==============================] - 1s - loss: 7.9049e-04 - acc: 0.8376 - val_loss: 0.0017 - val_acc: 0.7547 Epoch 1262/3000 1712/1712 [==============================] - 1s - loss: 8.6765e-04 - acc: 0.8370 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 1263/3000 1712/1712 [==============================] - 1s - loss: 7.7808e-04 - acc: 0.8259 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1264/3000 1712/1712 [==============================] - 1s - loss: 8.0394e-04 - acc: 0.8505 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 1265/3000 1712/1712 [==============================] - 1s - loss: 8.1179e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1266/3000 1712/1712 [==============================] - 1s - loss: 8.7419e-04 - acc: 0.8242 - val_loss: 0.0013 - val_acc: 0.7500 Epoch 1267/3000 1712/1712 [==============================] - 1s - loss: 7.5909e-04 - acc: 0.8318 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 1268/3000 1712/1712 [==============================] - 1s - loss: 8.3980e-04 - acc: 0.8294 - val_loss: 0.0014 - val_acc: 0.7266 Epoch 1269/3000 1712/1712 [==============================] - 1s - loss: 8.1241e-04 - acc: 0.8218 - val_loss: 0.0024 - val_acc: 0.7220 Epoch 1270/3000 1712/1712 [==============================] - 1s - loss: 8.7034e-04 - acc: 0.8370 - val_loss: 0.0022 - val_acc: 0.7079 Epoch 1271/3000 1712/1712 [==============================] - 1s - loss: 8.2084e-04 - acc: 0.8259 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1272/3000 1712/1712 [==============================] - 1s - loss: 7.7717e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 1273/3000 1712/1712 [==============================] - 1s - loss: 8.3498e-04 - acc: 0.8178 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 1274/3000 1712/1712 [==============================] - 1s - loss: 8.0958e-04 - acc: 0.8429 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 1275/3000 1712/1712 [==============================] - 1s - loss: 8.4900e-04 - acc: 0.8359 - val_loss: 0.0014 - val_acc: 0.7734 Epoch 1276/3000 1712/1712 [==============================] - 1s - loss: 7.9580e-04 - acc: 0.8324 - val_loss: 0.0020 - val_acc: 0.7196 Epoch 1277/3000 1712/1712 [==============================] - 1s - loss: 7.6168e-04 - acc: 0.8458 - val_loss: 0.0019 - val_acc: 0.7173 Epoch 1278/3000 1712/1712 [==============================] - 1s - loss: 8.5280e-04 - acc: 0.8254 - val_loss: 0.0017 - val_acc: 0.7664 Epoch 1279/3000 1712/1712 [==============================] - 1s - loss: 8.0926e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 1280/3000 1712/1712 [==============================] - 1s - loss: 8.2563e-04 - acc: 0.8318 - val_loss: 0.0012 - val_acc: 0.7523 Epoch 1281/3000 1712/1712 [==============================] - 1s - loss: 8.0410e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 1282/3000 1712/1712 [==============================] - 1s - loss: 7.9159e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1283/3000 1712/1712 [==============================] - 1s - loss: 8.8933e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 1284/3000 1712/1712 [==============================] - 1s - loss: 7.6828e-04 - acc: 0.8265 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 1285/3000 1712/1712 [==============================] - 1s - loss: 7.8746e-04 - acc: 0.8318 - val_loss: 0.0023 - val_acc: 0.7290 Epoch 1286/3000 1712/1712 [==============================] - 1s - loss: 8.6853e-04 - acc: 0.8254 - val_loss: 0.0019 - val_acc: 0.7827 Epoch 1287/3000 1712/1712 [==============================] - 1s - loss: 8.0637e-04 - acc: 0.8347 - val_loss: 0.0018 - val_acc: 0.7944 Epoch 1288/3000 1712/1712 [==============================] - 1s - loss: 8.6904e-04 - acc: 0.8189 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 1289/3000 1712/1712 [==============================] - 1s - loss: 8.0648e-04 - acc: 0.8440 - val_loss: 0.0020 - val_acc: 0.7103 Epoch 1290/3000 1712/1712 [==============================] - 1s - loss: 8.5734e-04 - acc: 0.8236 - val_loss: 0.0019 - val_acc: 0.7150 Epoch 1291/3000 1712/1712 [==============================] - 1s - loss: 7.8667e-04 - acc: 0.8271 - val_loss: 0.0015 - val_acc: 0.7570 Epoch 1292/3000 1712/1712 [==============================] - 1s - loss: 8.2424e-04 - acc: 0.8388 - val_loss: 0.0020 - val_acc: 0.7126 Epoch 1293/3000 1712/1712 [==============================] - 1s - loss: 7.1947e-04 - acc: 0.8376 - val_loss: 0.0018 - val_acc: 0.7290 Epoch 1294/3000 1712/1712 [==============================] - 1s - loss: 8.9716e-04 - acc: 0.8172 - val_loss: 0.0015 - val_acc: 0.7453 Epoch 1295/3000 1712/1712 [==============================] - 1s - loss: 8.2167e-04 - acc: 0.8242 - val_loss: 0.0024 - val_acc: 0.7313 Epoch 1296/3000 1712/1712 [==============================] - 1s - loss: 8.5615e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1297/3000 1712/1712 [==============================] - 1s - loss: 7.4659e-04 - acc: 0.8435 - val_loss: 0.0011 - val_acc: 0.8037 Epoch 1298/3000 1712/1712 [==============================] - 1s - loss: 8.5094e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1299/3000 1712/1712 [==============================] - 1s - loss: 8.4795e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.8201 Epoch 1300/3000 1712/1712 [==============================] - 1s - loss: 7.7197e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1301/3000 1712/1712 [==============================] - 1s - loss: 8.6925e-04 - acc: 0.8353 - val_loss: 0.0016 - val_acc: 0.7617 Epoch 1302/3000 1712/1712 [==============================] - 1s - loss: 7.5350e-04 - acc: 0.8335 - val_loss: 0.0021 - val_acc: 0.7360 Epoch 1303/3000 1712/1712 [==============================] - 1s - loss: 8.6032e-04 - acc: 0.8318 - val_loss: 0.0024 - val_acc: 0.7126 Epoch 1304/3000 1712/1712 [==============================] - 1s - loss: 7.9594e-04 - acc: 0.8306 - val_loss: 0.0019 - val_acc: 0.7173 Epoch 1305/3000 1712/1712 [==============================] - 1s - loss: 8.4585e-04 - acc: 0.8394 - val_loss: 0.0017 - val_acc: 0.7407 Epoch 1306/3000 1712/1712 [==============================] - 1s - loss: 7.7976e-04 - acc: 0.8271 - val_loss: 0.0020 - val_acc: 0.7477 Epoch 1307/3000 1712/1712 [==============================] - 1s - loss: 8.7294e-04 - acc: 0.8324 - val_loss: 0.0014 - val_acc: 0.7640 Epoch 1308/3000 1712/1712 [==============================] - 1s - loss: 8.2244e-04 - acc: 0.8312 - val_loss: 0.0015 - val_acc: 0.7477 Epoch 1309/3000 1712/1712 [==============================] - 1s - loss: 8.6801e-04 - acc: 0.8224 - val_loss: 0.0017 - val_acc: 0.7547 Epoch 1310/3000 1712/1712 [==============================] - 1s - loss: 7.7669e-04 - acc: 0.8458 - val_loss: 0.0025 - val_acc: 0.7079 Epoch 1311/3000 1712/1712 [==============================] - 1s - loss: 8.5970e-04 - acc: 0.8201 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 1312/3000 1712/1712 [==============================] - 1s - loss: 8.4419e-04 - acc: 0.8160 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1313/3000 1712/1712 [==============================] - 1s - loss: 8.4955e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 1314/3000 1712/1712 [==============================] - 1s - loss: 8.3082e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1315/3000 1712/1712 [==============================] - 1s - loss: 7.1800e-04 - acc: 0.8353 - val_loss: 0.0011 - val_acc: 0.8037 Epoch 1316/3000 1712/1712 [==============================] - 1s - loss: 9.2742e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 1317/3000 1712/1712 [==============================] - 1s - loss: 7.1265e-04 - acc: 0.8452 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1318/3000 1712/1712 [==============================] - 1s - loss: 9.1837e-04 - acc: 0.8435 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 1319/3000 1712/1712 [==============================] - 1s - loss: 8.0705e-04 - acc: 0.8248 - val_loss: 0.0016 - val_acc: 0.7874 Epoch 1320/3000 1712/1712 [==============================] - 1s - loss: 7.6919e-04 - acc: 0.8347 - val_loss: 0.0015 - val_acc: 0.7921 Epoch 1321/3000 1712/1712 [==============================] - 1s - loss: 8.0149e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.8224 Epoch 1322/3000 1712/1712 [==============================] - 1s - loss: 8.2856e-04 - acc: 0.8440 - val_loss: 0.0011 - val_acc: 0.8084 Epoch 1323/3000 1712/1712 [==============================] - 1s - loss: 8.0479e-04 - acc: 0.8452 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 1324/3000 1712/1712 [==============================] - 1s - loss: 8.3038e-04 - acc: 0.8329 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 1325/3000 1712/1712 [==============================] - 1s - loss: 7.4536e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1326/3000 1712/1712 [==============================] - 1s - loss: 8.4768e-04 - acc: 0.8394 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 1327/3000 1712/1712 [==============================] - 1s - loss: 8.2662e-04 - acc: 0.8137 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 1328/3000 1712/1712 [==============================] - 1s - loss: 8.3351e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 1329/3000 1712/1712 [==============================] - 1s - loss: 7.9467e-04 - acc: 0.8254 - val_loss: 0.0014 - val_acc: 0.7290 Epoch 1330/3000 1712/1712 [==============================] - 1s - loss: 8.8234e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 1331/3000 1712/1712 [==============================] - 1s - loss: 7.5918e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1332/3000 1712/1712 [==============================] - 1s - loss: 8.7069e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 1333/3000 1712/1712 [==============================] - 1s - loss: 8.2249e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1334/3000 1712/1712 [==============================] - 1s - loss: 7.9144e-04 - acc: 0.8160 - val_loss: 0.0011 - val_acc: 0.7757 Epoch 1335/3000 1712/1712 [==============================] - 1s - loss: 8.2029e-04 - acc: 0.8464 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 1336/3000 1712/1712 [==============================] - 1s - loss: 8.5582e-04 - acc: 0.8294 - val_loss: 0.0014 - val_acc: 0.7640 Epoch 1337/3000 1712/1712 [==============================] - 1s - loss: 8.0960e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 1338/3000 1712/1712 [==============================] - 1s - loss: 7.5515e-04 - acc: 0.8394 - val_loss: 0.0021 - val_acc: 0.7313 Epoch 1339/3000 1712/1712 [==============================] - 1s - loss: 8.7577e-04 - acc: 0.8341 - val_loss: 0.0018 - val_acc: 0.7407 Epoch 1340/3000 1712/1712 [==============================] - 1s - loss: 7.8482e-04 - acc: 0.8329 - val_loss: 0.0016 - val_acc: 0.7220 Epoch 1341/3000 1712/1712 [==============================] - 1s - loss: 8.3145e-04 - acc: 0.8178 - val_loss: 0.0019 - val_acc: 0.7430 Epoch 1342/3000 1712/1712 [==============================] - 1s - loss: 7.6142e-04 - acc: 0.8294 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 1343/3000 1712/1712 [==============================] - 1s - loss: 8.8830e-04 - acc: 0.8300 - val_loss: 0.0016 - val_acc: 0.7617 Epoch 1344/3000 1712/1712 [==============================] - 1s - loss: 8.7025e-04 - acc: 0.8294 - val_loss: 0.0015 - val_acc: 0.7710 Epoch 1345/3000 1712/1712 [==============================] - 1s - loss: 7.5317e-04 - acc: 0.8405 - val_loss: 0.0018 - val_acc: 0.7360 Epoch 1346/3000 1712/1712 [==============================] - 1s - loss: 9.0457e-04 - acc: 0.8341 - val_loss: 0.0019 - val_acc: 0.7150 Epoch 1347/3000 1712/1712 [==============================] - 1s - loss: 7.8846e-04 - acc: 0.8435 - val_loss: 0.0019 - val_acc: 0.7430 Epoch 1348/3000 1712/1712 [==============================] - 1s - loss: 8.0640e-04 - acc: 0.8470 - val_loss: 0.0018 - val_acc: 0.7827 Epoch 1349/3000 1712/1712 [==============================] - 1s - loss: 9.1684e-04 - acc: 0.8417 - val_loss: 0.0019 - val_acc: 0.7687 Epoch 1350/3000 1712/1712 [==============================] - 1s - loss: 7.1951e-04 - acc: 0.8458 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 1351/3000 1712/1712 [==============================] - 1s - loss: 8.5466e-04 - acc: 0.8289 - val_loss: 0.0011 - val_acc: 0.8248 Epoch 1352/3000 1712/1712 [==============================] - 1s - loss: 7.9519e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 1353/3000 1712/1712 [==============================] - 1s - loss: 8.1531e-04 - acc: 0.8218 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 1354/3000 1712/1712 [==============================] - 1s - loss: 8.6673e-04 - acc: 0.8224 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 1355/3000 1712/1712 [==============================] - 1s - loss: 7.3267e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1356/3000 1712/1712 [==============================] - 1s - loss: 8.2568e-04 - acc: 0.8382 - val_loss: 0.0012 - val_acc: 0.8294 Epoch 1357/3000 1712/1712 [==============================] - 1s - loss: 9.0931e-04 - acc: 0.8183 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 1358/3000 1712/1712 [==============================] - 1s - loss: 7.4942e-04 - acc: 0.8259 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1359/3000 1712/1712 [==============================] - 1s - loss: 8.4951e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1360/3000 1712/1712 [==============================] - 1s - loss: 8.0858e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.7430 Epoch 1361/3000 1712/1712 [==============================] - 1s - loss: 7.4363e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1362/3000 1712/1712 [==============================] - 1s - loss: 8.8839e-04 - acc: 0.8353 - val_loss: 0.0015 - val_acc: 0.7617 Epoch 1363/3000 1712/1712 [==============================] - 1s - loss: 8.3444e-04 - acc: 0.8289 - val_loss: 0.0019 - val_acc: 0.7500 Epoch 1364/3000 1712/1712 [==============================] - 1s - loss: 8.0908e-04 - acc: 0.8452 - val_loss: 0.0014 - val_acc: 0.7850 Epoch 1365/3000 1712/1712 [==============================] - 1s - loss: 8.2114e-04 - acc: 0.8154 - val_loss: 0.0012 - val_acc: 0.7523 Epoch 1366/3000 1712/1712 [==============================] - 1s - loss: 8.7109e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 1367/3000 1712/1712 [==============================] - 1s - loss: 7.8615e-04 - acc: 0.8405 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1368/3000 1712/1712 [==============================] - 1s - loss: 8.1946e-04 - acc: 0.8259 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1369/3000 1712/1712 [==============================] - 1s - loss: 8.5834e-04 - acc: 0.8259 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 1370/3000 1712/1712 [==============================] - 1s - loss: 7.3449e-04 - acc: 0.8324 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 1371/3000 1712/1712 [==============================] - 1s - loss: 8.2762e-04 - acc: 0.8394 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 1372/3000 1712/1712 [==============================] - 1s - loss: 7.7401e-04 - acc: 0.8464 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 1373/3000 1712/1712 [==============================] - 1s - loss: 8.6501e-04 - acc: 0.8318 - val_loss: 0.0014 - val_acc: 0.7523 Epoch 1374/3000 1712/1712 [==============================] - 1s - loss: 7.5893e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 1375/3000 1712/1712 [==============================] - 1s - loss: 7.8546e-04 - acc: 0.8475 - val_loss: 0.0011 - val_acc: 0.7827 Epoch 1376/3000 1712/1712 [==============================] - 1s - loss: 8.7501e-04 - acc: 0.8353 - val_loss: 0.0014 - val_acc: 0.7757 Epoch 1377/3000 1712/1712 [==============================] - 1s - loss: 7.7826e-04 - acc: 0.8470 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 1378/3000 1712/1712 [==============================] - 1s - loss: 8.6933e-04 - acc: 0.8294 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 1379/3000 1712/1712 [==============================] - 1s - loss: 7.5661e-04 - acc: 0.8359 - val_loss: 0.0020 - val_acc: 0.7173 Epoch 1380/3000 1712/1712 [==============================] - 1s - loss: 7.9664e-04 - acc: 0.8341 - val_loss: 0.0020 - val_acc: 0.7103 Epoch 1381/3000 1712/1712 [==============================] - 1s - loss: 8.3675e-04 - acc: 0.8364 - val_loss: 0.0015 - val_acc: 0.7430 Epoch 1382/3000 1712/1712 [==============================] - 1s - loss: 7.8249e-04 - acc: 0.8265 - val_loss: 0.0020 - val_acc: 0.7313 Epoch 1383/3000 1712/1712 [==============================] - 1s - loss: 7.9215e-04 - acc: 0.8277 - val_loss: 0.0017 - val_acc: 0.7383 Epoch 1384/3000 1712/1712 [==============================] - 1s - loss: 7.8815e-04 - acc: 0.8312 - val_loss: 0.0022 - val_acc: 0.7313 Epoch 1385/3000 1712/1712 [==============================] - 1s - loss: 7.8254e-04 - acc: 0.8318 - val_loss: 0.0012 - val_acc: 0.7523 Epoch 1386/3000 1712/1712 [==============================] - 1s - loss: 9.0297e-04 - acc: 0.8242 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 1387/3000 1712/1712 [==============================] - 1s - loss: 7.0326e-04 - acc: 0.8505 - val_loss: 0.0011 - val_acc: 0.8061 Epoch 1388/3000 1712/1712 [==============================] - 1s - loss: 8.4458e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1389/3000 1712/1712 [==============================] - 1s - loss: 8.1084e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1390/3000 1712/1712 [==============================] - 1s - loss: 7.9773e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 1391/3000 1712/1712 [==============================] - 1s - loss: 7.8454e-04 - acc: 0.8359 - val_loss: 0.0011 - val_acc: 0.8224 Epoch 1392/3000 1712/1712 [==============================] - 1s - loss: 8.2868e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.8154 Epoch 1393/3000 1712/1712 [==============================] - 1s - loss: 8.6266e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1394/3000 1712/1712 [==============================] - 1s - loss: 7.9865e-04 - acc: 0.8207 - val_loss: 0.0015 - val_acc: 0.7874 Epoch 1395/3000 1712/1712 [==============================] - 1s - loss: 8.5472e-04 - acc: 0.8429 - val_loss: 0.0015 - val_acc: 0.7453 Epoch 1396/3000 1712/1712 [==============================] - 1s - loss: 8.2655e-04 - acc: 0.8218 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 1397/3000 1712/1712 [==============================] - 1s - loss: 7.7721e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 1398/3000 1712/1712 [==============================] - 1s - loss: 7.9341e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1399/3000 1712/1712 [==============================] - 1s - loss: 8.4445e-04 - acc: 0.8324 - val_loss: 0.0014 - val_acc: 0.7967 Epoch 1400/3000 1712/1712 [==============================] - 1s - loss: 7.8407e-04 - acc: 0.8382 - val_loss: 0.0011 - val_acc: 0.7944 Epoch 1401/3000 1712/1712 [==============================] - 1s - loss: 8.2571e-04 - acc: 0.8359 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 1402/3000 1712/1712 [==============================] - 1s - loss: 7.9452e-04 - acc: 0.8160 - val_loss: 0.0011 - val_acc: 0.8248 Epoch 1403/3000 1712/1712 [==============================] - 1s - loss: 8.2889e-04 - acc: 0.8300 - val_loss: 0.0014 - val_acc: 0.8014 Epoch 1404/3000 1712/1712 [==============================] - 1s - loss: 7.9611e-04 - acc: 0.8335 - val_loss: 0.0014 - val_acc: 0.7757 Epoch 1405/3000 1712/1712 [==============================] - 1s - loss: 8.4476e-04 - acc: 0.8411 - val_loss: 0.0012 - val_acc: 0.7640 Epoch 1406/3000 1712/1712 [==============================] - 1s - loss: 7.6709e-04 - acc: 0.8400 - val_loss: 0.0012 - val_acc: 0.8178 Epoch 1407/3000 1712/1712 [==============================] - 1s - loss: 8.2269e-04 - acc: 0.8370 - val_loss: 0.0011 - val_acc: 0.8248 Epoch 1408/3000 1712/1712 [==============================] - 1s - loss: 8.0736e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7407 Epoch 1409/3000 1712/1712 [==============================] - 1s - loss: 8.4685e-04 - acc: 0.8341 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 1410/3000 1712/1712 [==============================] - 1s - loss: 7.9175e-04 - acc: 0.8341 - val_loss: 0.0014 - val_acc: 0.7523 Epoch 1411/3000 1712/1712 [==============================] - 1s - loss: 8.0810e-04 - acc: 0.8435 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1412/3000 1712/1712 [==============================] - 1s - loss: 6.6908e-04 - acc: 0.8341 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 1413/3000 1712/1712 [==============================] - 1s - loss: 8.7448e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7547 Epoch 1414/3000 1712/1712 [==============================] - 1s - loss: 8.9958e-04 - acc: 0.8242 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 1415/3000 1712/1712 [==============================] - 1s - loss: 8.0324e-04 - acc: 0.8440 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 1416/3000 1712/1712 [==============================] - 1s - loss: 8.3233e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1417/3000 1712/1712 [==============================] - 1s - loss: 7.7202e-04 - acc: 0.8464 - val_loss: 0.0021 - val_acc: 0.7477 Epoch 1418/3000 1712/1712 [==============================] - 1s - loss: 7.9609e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1419/3000 1712/1712 [==============================] - 1s - loss: 8.1627e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1420/3000 1712/1712 [==============================] - 1s - loss: 8.1356e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 1421/3000 1712/1712 [==============================] - 1s - loss: 8.1099e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1422/3000 1712/1712 [==============================] - 1s - loss: 7.7162e-04 - acc: 0.8475 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1423/3000 1712/1712 [==============================] - 1s - loss: 8.0736e-04 - acc: 0.8429 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 1424/3000 1712/1712 [==============================] - 1s - loss: 8.0998e-04 - acc: 0.8364 - val_loss: 0.0014 - val_acc: 0.7407 Epoch 1425/3000 1712/1712 [==============================] - 1s - loss: 7.7056e-04 - acc: 0.8417 - val_loss: 0.0012 - val_acc: 0.7664 Epoch 1426/3000 1712/1712 [==============================] - 1s - loss: 8.5442e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7640 Epoch 1427/3000 1712/1712 [==============================] - 1s - loss: 7.8984e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1428/3000 1712/1712 [==============================] - 1s - loss: 8.2615e-04 - acc: 0.8546 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 1429/3000 1712/1712 [==============================] - 1s - loss: 7.7114e-04 - acc: 0.8254 - val_loss: 0.0014 - val_acc: 0.7523 Epoch 1430/3000 1712/1712 [==============================] - 1s - loss: 7.8644e-04 - acc: 0.8347 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 1431/3000 1712/1712 [==============================] - 1s - loss: 8.3617e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1432/3000 1712/1712 [==============================] - 1s - loss: 8.0003e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.8061 Epoch 1433/3000 1712/1712 [==============================] - 1s - loss: 8.0106e-04 - acc: 0.8213 - val_loss: 0.0011 - val_acc: 0.8178 Epoch 1434/3000 1712/1712 [==============================] - 1s - loss: 7.9520e-04 - acc: 0.8470 - val_loss: 0.0011 - val_acc: 0.8014 Epoch 1435/3000 1712/1712 [==============================] - 1s - loss: 8.7114e-04 - acc: 0.8160 - val_loss: 0.0012 - val_acc: 0.7547 Epoch 1436/3000 1712/1712 [==============================] - 1s - loss: 7.4731e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 1437/3000 1712/1712 [==============================] - 1s - loss: 8.5756e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.7547 Epoch 1438/3000 1712/1712 [==============================] - 1s - loss: 8.1246e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 1439/3000 1712/1712 [==============================] - 1s - loss: 8.0572e-04 - acc: 0.8236 - val_loss: 0.0016 - val_acc: 0.7407 Epoch 1440/3000 1712/1712 [==============================] - 1s - loss: 8.4787e-04 - acc: 0.8400 - val_loss: 0.0020 - val_acc: 0.7710 Epoch 1441/3000 1712/1712 [==============================] - 1s - loss: 8.1252e-04 - acc: 0.8254 - val_loss: 0.0020 - val_acc: 0.7874 Epoch 1442/3000 1712/1712 [==============================] - 1s - loss: 8.0469e-04 - acc: 0.8364 - val_loss: 0.0016 - val_acc: 0.7360 Epoch 1443/3000 1712/1712 [==============================] - 1s - loss: 9.2991e-04 - acc: 0.8224 - val_loss: 0.0017 - val_acc: 0.7430 Epoch 1444/3000 1712/1712 [==============================] - 1s - loss: 7.6937e-04 - acc: 0.8294 - val_loss: 0.0016 - val_acc: 0.7734 Epoch 1445/3000 1712/1712 [==============================] - 1s - loss: 8.1142e-04 - acc: 0.8505 - val_loss: 0.0011 - val_acc: 0.7991 Epoch 1446/3000 1712/1712 [==============================] - 1s - loss: 7.7820e-04 - acc: 0.8324 - val_loss: 0.0016 - val_acc: 0.7243 Epoch 1447/3000 1712/1712 [==============================] - 1s - loss: 8.6651e-04 - acc: 0.8300 - val_loss: 0.0017 - val_acc: 0.7570 Epoch 1448/3000 1712/1712 [==============================] - 1s - loss: 7.6049e-04 - acc: 0.8242 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 1449/3000 1712/1712 [==============================] - 1s - loss: 8.4231e-04 - acc: 0.8271 - val_loss: 0.0019 - val_acc: 0.7150 Epoch 1450/3000 1712/1712 [==============================] - 1s - loss: 8.0859e-04 - acc: 0.8242 - val_loss: 0.0019 - val_acc: 0.7383 Epoch 1451/3000 1712/1712 [==============================] - 1s - loss: 7.7385e-04 - acc: 0.8400 - val_loss: 0.0027 - val_acc: 0.7126 Epoch 1452/3000 1712/1712 [==============================] - 1s - loss: 9.0835e-04 - acc: 0.8224 - val_loss: 0.0021 - val_acc: 0.7033 Epoch 1453/3000 1712/1712 [==============================] - 1s - loss: 7.5697e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1454/3000 1712/1712 [==============================] - 1s - loss: 8.5140e-04 - acc: 0.8271 - val_loss: 0.0011 - val_acc: 0.8037 Epoch 1455/3000 1712/1712 [==============================] - 1s - loss: 7.9457e-04 - acc: 0.8318 - val_loss: 0.0015 - val_acc: 0.7477 Epoch 1456/3000 1712/1712 [==============================] - 1s - loss: 8.7938e-04 - acc: 0.8201 - val_loss: 0.0019 - val_acc: 0.7290 Epoch 1457/3000 1712/1712 [==============================] - 1s - loss: 7.0596e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.8271 Epoch 1458/3000 1712/1712 [==============================] - 1s - loss: 9.3291e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1459/3000 1712/1712 [==============================] - 1s - loss: 7.9831e-04 - acc: 0.8446 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 1460/3000 1712/1712 [==============================] - 1s - loss: 8.3163e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 1461/3000 1712/1712 [==============================] - 1s - loss: 8.2216e-04 - acc: 0.8353 - val_loss: 0.0014 - val_acc: 0.7500 Epoch 1462/3000 1712/1712 [==============================] - 1s - loss: 7.7738e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 1463/3000 1712/1712 [==============================] - 1s - loss: 8.2246e-04 - acc: 0.8359 - val_loss: 0.0015 - val_acc: 0.7383 Epoch 1464/3000 1712/1712 [==============================] - 1s - loss: 8.0686e-04 - acc: 0.8195 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 1465/3000 1712/1712 [==============================] - 1s - loss: 7.5670e-04 - acc: 0.8429 - val_loss: 0.0019 - val_acc: 0.7500 Epoch 1466/3000 1712/1712 [==============================] - 1s - loss: 7.5757e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 1467/3000 1712/1712 [==============================] - 1s - loss: 8.5233e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 1468/3000 1712/1712 [==============================] - 1s - loss: 7.9476e-04 - acc: 0.8318 - val_loss: 0.0012 - val_acc: 0.7570 Epoch 1469/3000 1712/1712 [==============================] - 1s - loss: 8.1025e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 1470/3000 1712/1712 [==============================] - 1s - loss: 8.4028e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 1471/3000 1712/1712 [==============================] - 1s - loss: 8.0523e-04 - acc: 0.8230 - val_loss: 0.0014 - val_acc: 0.7477 Epoch 1472/3000 1712/1712 [==============================] - 1s - loss: 8.4135e-04 - acc: 0.8236 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 1473/3000 1712/1712 [==============================] - 1s - loss: 7.5690e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1474/3000 1712/1712 [==============================] - 1s - loss: 9.1487e-04 - acc: 0.8125 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 1475/3000 1712/1712 [==============================] - 1s - loss: 7.5547e-04 - acc: 0.8388 - val_loss: 0.0024 - val_acc: 0.7173 Epoch 1476/3000 1712/1712 [==============================] - 1s - loss: 8.0096e-04 - acc: 0.8405 - val_loss: 0.0014 - val_acc: 0.8154 Epoch 1477/3000 1712/1712 [==============================] - 1s - loss: 8.9514e-04 - acc: 0.8242 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1478/3000 1712/1712 [==============================] - 1s - loss: 7.6239e-04 - acc: 0.8423 - val_loss: 0.0013 - val_acc: 0.8201 Epoch 1479/3000 1712/1712 [==============================] - 1s - loss: 8.0604e-04 - acc: 0.8347 - val_loss: 0.0011 - val_acc: 0.8201 Epoch 1480/3000 1712/1712 [==============================] - 1s - loss: 7.3901e-04 - acc: 0.8394 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 1481/3000 1712/1712 [==============================] - 1s - loss: 9.2056e-04 - acc: 0.8043 - val_loss: 0.0011 - val_acc: 0.7944 Epoch 1482/3000 1712/1712 [==============================] - 1s - loss: 7.2692e-04 - acc: 0.8440 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 1483/3000 1712/1712 [==============================] - 1s - loss: 8.2463e-04 - acc: 0.8370 - val_loss: 0.0011 - val_acc: 0.7944 Epoch 1484/3000 1712/1712 [==============================] - 1s - loss: 8.2836e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 1485/3000 1712/1712 [==============================] - 1s - loss: 7.8085e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 1486/3000 1712/1712 [==============================] - 1s - loss: 8.3411e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 1487/3000 1712/1712 [==============================] - 1s - loss: 8.1991e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1488/3000 1712/1712 [==============================] - 1s - loss: 7.9836e-04 - acc: 0.8364 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 1489/3000 1712/1712 [==============================] - 1s - loss: 7.9437e-04 - acc: 0.8364 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 1490/3000 1712/1712 [==============================] - 1s - loss: 7.8305e-04 - acc: 0.8370 - val_loss: 0.0016 - val_acc: 0.7850 Epoch 1491/3000 1712/1712 [==============================] - 1s - loss: 8.4303e-04 - acc: 0.8347 - val_loss: 0.0021 - val_acc: 0.7523 Epoch 1492/3000 1712/1712 [==============================] - 1s - loss: 8.6586e-04 - acc: 0.8236 - val_loss: 0.0016 - val_acc: 0.7383 Epoch 1493/3000 1712/1712 [==============================] - 1s - loss: 8.6296e-04 - acc: 0.8376 - val_loss: 0.0020 - val_acc: 0.7477 Epoch 1494/3000 1712/1712 [==============================] - 1s - loss: 7.6574e-04 - acc: 0.8376 - val_loss: 0.0018 - val_acc: 0.7734 Epoch 1495/3000 1712/1712 [==============================] - 1s - loss: 7.5607e-04 - acc: 0.8376 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 1496/3000 1712/1712 [==============================] - 1s - loss: 9.1068e-04 - acc: 0.8107 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 1497/3000 1712/1712 [==============================] - 1s - loss: 7.4914e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1498/3000 1712/1712 [==============================] - 1s - loss: 8.2080e-04 - acc: 0.8324 - val_loss: 0.0020 - val_acc: 0.7383 Epoch 1499/3000 1712/1712 [==============================] - 1s - loss: 8.1652e-04 - acc: 0.8300 - val_loss: 0.0020 - val_acc: 0.7290 Epoch 1500/3000 1712/1712 [==============================] - 1s - loss: 8.6607e-04 - acc: 0.8306 - val_loss: 0.0016 - val_acc: 0.7593 Epoch 1501/3000 1712/1712 [==============================] - 1s - loss: 8.2062e-04 - acc: 0.8353 - val_loss: 0.0020 - val_acc: 0.7640 Epoch 1502/3000 1712/1712 [==============================] - 1s - loss: 8.1697e-04 - acc: 0.8289 - val_loss: 0.0020 - val_acc: 0.7336 Epoch 1503/3000 1712/1712 [==============================] - 1s - loss: 7.7187e-04 - acc: 0.8417 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 1504/3000 1712/1712 [==============================] - 1s - loss: 8.1463e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 1505/3000 1712/1712 [==============================] - 1s - loss: 7.2907e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 1506/3000 1712/1712 [==============================] - 1s - loss: 8.0532e-04 - acc: 0.8324 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 1507/3000 1712/1712 [==============================] - 1s - loss: 8.4656e-04 - acc: 0.8183 - val_loss: 0.0013 - val_acc: 0.7593 Epoch 1508/3000 1712/1712 [==============================] - 1s - loss: 8.1905e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 1509/3000 1712/1712 [==============================] - 1s - loss: 8.1578e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 1510/3000 1712/1712 [==============================] - 1s - loss: 8.2686e-04 - acc: 0.8329 - val_loss: 0.0014 - val_acc: 0.7874 Epoch 1511/3000 1712/1712 [==============================] - 1s - loss: 8.3958e-04 - acc: 0.8394 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1512/3000 1712/1712 [==============================] - 1s - loss: 6.9922e-04 - acc: 0.8370 - val_loss: 0.0014 - val_acc: 0.7991 Epoch 1513/3000 1712/1712 [==============================] - 1s - loss: 8.8685e-04 - acc: 0.8370 - val_loss: 0.0017 - val_acc: 0.7640 Epoch 1514/3000 1712/1712 [==============================] - 1s - loss: 7.7491e-04 - acc: 0.8411 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 1515/3000 1712/1712 [==============================] - 1s - loss: 8.0109e-04 - acc: 0.8423 - val_loss: 0.0014 - val_acc: 0.7477 Epoch 1516/3000 1712/1712 [==============================] - 1s - loss: 7.8855e-04 - acc: 0.8312 - val_loss: 0.0019 - val_acc: 0.7196 Epoch 1517/3000 1712/1712 [==============================] - 1s - loss: 8.2526e-04 - acc: 0.8318 - val_loss: 0.0016 - val_acc: 0.7523 Epoch 1518/3000 1712/1712 [==============================] - 1s - loss: 8.7443e-04 - acc: 0.8405 - val_loss: 0.0022 - val_acc: 0.7079 Epoch 1519/3000 1712/1712 [==============================] - 1s - loss: 8.0259e-04 - acc: 0.8318 - val_loss: 0.0024 - val_acc: 0.7033 Epoch 1520/3000 1712/1712 [==============================] - 1s - loss: 7.5623e-04 - acc: 0.8405 - val_loss: 0.0016 - val_acc: 0.7500 Epoch 1521/3000 1712/1712 [==============================] - 1s - loss: 8.1794e-04 - acc: 0.8242 - val_loss: 0.0012 - val_acc: 0.7664 Epoch 1522/3000 1712/1712 [==============================] - 1s - loss: 8.5983e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1523/3000 1712/1712 [==============================] - 1s - loss: 7.6153e-04 - acc: 0.8277 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 1524/3000 1712/1712 [==============================] - 1s - loss: 8.7659e-04 - acc: 0.8259 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1525/3000 1712/1712 [==============================] - 1s - loss: 8.1191e-04 - acc: 0.8370 - val_loss: 0.0015 - val_acc: 0.7617 Epoch 1526/3000 1712/1712 [==============================] - 1s - loss: 7.8336e-04 - acc: 0.8289 - val_loss: 0.0018 - val_acc: 0.7804 Epoch 1527/3000 1712/1712 [==============================] - 1s - loss: 7.9264e-04 - acc: 0.8435 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 1528/3000 1712/1712 [==============================] - 1s - loss: 7.7798e-04 - acc: 0.8394 - val_loss: 0.0018 - val_acc: 0.7547 Epoch 1529/3000 1712/1712 [==============================] - 1s - loss: 8.7708e-04 - acc: 0.8271 - val_loss: 0.0020 - val_acc: 0.7196 Epoch 1530/3000 1712/1712 [==============================] - 1s - loss: 7.8505e-04 - acc: 0.8446 - val_loss: 0.0016 - val_acc: 0.7313 Epoch 1531/3000 1712/1712 [==============================] - 1s - loss: 8.2578e-04 - acc: 0.8242 - val_loss: 0.0016 - val_acc: 0.7477 Epoch 1532/3000 1712/1712 [==============================] - 1s - loss: 8.7718e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 1533/3000 1712/1712 [==============================] - 1s - loss: 8.4448e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 1534/3000 1712/1712 [==============================] - 1s - loss: 7.9416e-04 - acc: 0.8283 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 1535/3000 1712/1712 [==============================] - 1s - loss: 7.6736e-04 - acc: 0.8458 - val_loss: 0.0016 - val_acc: 0.7617 Epoch 1536/3000 1712/1712 [==============================] - 1s - loss: 8.1692e-04 - acc: 0.8189 - val_loss: 0.0015 - val_acc: 0.8014 Epoch 1537/3000 1712/1712 [==============================] - 1s - loss: 8.7587e-04 - acc: 0.8300 - val_loss: 0.0017 - val_acc: 0.7617 Epoch 1538/3000 1712/1712 [==============================] - 1s - loss: 7.8459e-04 - acc: 0.8376 - val_loss: 0.0014 - val_acc: 0.7570 Epoch 1539/3000 1712/1712 [==============================] - 1s - loss: 8.1197e-04 - acc: 0.8347 - val_loss: 0.0014 - val_acc: 0.7874 Epoch 1540/3000 1712/1712 [==============================] - 1s - loss: 7.9592e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 1541/3000 1712/1712 [==============================] - 1s - loss: 8.2374e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 1542/3000 1712/1712 [==============================] - 1s - loss: 8.1829e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7477 Epoch 1543/3000 1712/1712 [==============================] - 1s - loss: 8.7996e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 1544/3000 1712/1712 [==============================] - 1s - loss: 7.8251e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1545/3000 1712/1712 [==============================] - 1s - loss: 7.8268e-04 - acc: 0.8329 - val_loss: 0.0019 - val_acc: 0.7079 Epoch 1546/3000 1712/1712 [==============================] - 1s - loss: 8.1747e-04 - acc: 0.8306 - val_loss: 0.0019 - val_acc: 0.7150 Epoch 1547/3000 1712/1712 [==============================] - 1s - loss: 8.8858e-04 - acc: 0.8189 - val_loss: 0.0018 - val_acc: 0.7360 Epoch 1548/3000 1712/1712 [==============================] - 1s - loss: 8.5103e-04 - acc: 0.8294 - val_loss: 0.0018 - val_acc: 0.7827 Epoch 1549/3000 1712/1712 [==============================] - 1s - loss: 7.4895e-04 - acc: 0.8341 - val_loss: 0.0019 - val_acc: 0.7547 Epoch 1550/3000 1712/1712 [==============================] - 1s - loss: 8.3808e-04 - acc: 0.8329 - val_loss: 0.0017 - val_acc: 0.7547 Epoch 1551/3000 1712/1712 [==============================] - 1s - loss: 8.2619e-04 - acc: 0.8388 - val_loss: 0.0014 - val_acc: 0.8037 Epoch 1552/3000 1712/1712 [==============================] - 1s - loss: 7.8013e-04 - acc: 0.8440 - val_loss: 0.0018 - val_acc: 0.7430 Epoch 1553/3000 1712/1712 [==============================] - 1s - loss: 8.3320e-04 - acc: 0.8335 - val_loss: 0.0019 - val_acc: 0.7173 Epoch 1554/3000 1712/1712 [==============================] - 1s - loss: 7.7892e-04 - acc: 0.8166 - val_loss: 0.0025 - val_acc: 0.7173 Epoch 1555/3000 1712/1712 [==============================] - 1s - loss: 8.6870e-04 - acc: 0.8318 - val_loss: 0.0019 - val_acc: 0.7664 Epoch 1556/3000 1712/1712 [==============================] - 1s - loss: 8.0535e-04 - acc: 0.8353 - val_loss: 0.0018 - val_acc: 0.7500 Epoch 1557/3000 1712/1712 [==============================] - 1s - loss: 7.9054e-04 - acc: 0.8300 - val_loss: 0.0019 - val_acc: 0.7617 Epoch 1558/3000 1712/1712 [==============================] - 1s - loss: 8.3271e-04 - acc: 0.8213 - val_loss: 0.0017 - val_acc: 0.7640 Epoch 1559/3000 1712/1712 [==============================] - 1s - loss: 7.4088e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7593 Epoch 1560/3000 1712/1712 [==============================] - 1s - loss: 9.4956e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 1561/3000 1712/1712 [==============================] - 1s - loss: 7.4173e-04 - acc: 0.8300 - val_loss: 0.0019 - val_acc: 0.7570 Epoch 1562/3000 1712/1712 [==============================] - 1s - loss: 8.5414e-04 - acc: 0.8318 - val_loss: 0.0018 - val_acc: 0.7173 Epoch 1563/3000 1712/1712 [==============================] - 1s - loss: 7.5381e-04 - acc: 0.8259 - val_loss: 0.0025 - val_acc: 0.7079 Epoch 1564/3000 1712/1712 [==============================] - 1s - loss: 8.4521e-04 - acc: 0.8312 - val_loss: 0.0017 - val_acc: 0.7827 Epoch 1565/3000 1712/1712 [==============================] - 1s - loss: 8.4821e-04 - acc: 0.8394 - val_loss: 0.0024 - val_acc: 0.7243 Epoch 1566/3000 1712/1712 [==============================] - 1s - loss: 8.3470e-04 - acc: 0.8300 - val_loss: 0.0019 - val_acc: 0.7383 Epoch 1567/3000 1712/1712 [==============================] - 1s - loss: 7.9242e-04 - acc: 0.8072 - val_loss: 0.0022 - val_acc: 0.7103 Epoch 1568/3000 1712/1712 [==============================] - 1s - loss: 7.7275e-04 - acc: 0.8376 - val_loss: 0.0017 - val_acc: 0.7897 Epoch 1569/3000 1712/1712 [==============================] - 1s - loss: 8.9071e-04 - acc: 0.8324 - val_loss: 0.0020 - val_acc: 0.7266 Epoch 1570/3000 1712/1712 [==============================] - 1s - loss: 7.6524e-04 - acc: 0.8382 - val_loss: 0.0017 - val_acc: 0.7687 Epoch 1571/3000 1712/1712 [==============================] - 1s - loss: 8.2642e-04 - acc: 0.8359 - val_loss: 0.0016 - val_acc: 0.7477 Epoch 1572/3000 1712/1712 [==============================] - 1s - loss: 8.2449e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 1573/3000 1712/1712 [==============================] - 1s - loss: 7.3706e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 1574/3000 1712/1712 [==============================] - 1s - loss: 8.7110e-04 - acc: 0.8324 - val_loss: 0.0020 - val_acc: 0.7570 Epoch 1575/3000 1712/1712 [==============================] - 1s - loss: 8.0452e-04 - acc: 0.8394 - val_loss: 0.0020 - val_acc: 0.7336 Epoch 1576/3000 1712/1712 [==============================] - 1s - loss: 7.8238e-04 - acc: 0.8283 - val_loss: 0.0017 - val_acc: 0.7897 Epoch 1577/3000 1712/1712 [==============================] - 1s - loss: 8.5547e-04 - acc: 0.8254 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 1578/3000 1712/1712 [==============================] - 1s - loss: 7.4309e-04 - acc: 0.8417 - val_loss: 0.0014 - val_acc: 0.7944 Epoch 1579/3000 1712/1712 [==============================] - 1s - loss: 9.0949e-04 - acc: 0.8341 - val_loss: 0.0014 - val_acc: 0.7453 Epoch 1580/3000 1712/1712 [==============================] - 1s - loss: 8.6086e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1581/3000 1712/1712 [==============================] - 1s - loss: 7.7726e-04 - acc: 0.8452 - val_loss: 0.0015 - val_acc: 0.7383 Epoch 1582/3000 1712/1712 [==============================] - 1s - loss: 8.0681e-04 - acc: 0.8265 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 1583/3000 1712/1712 [==============================] - 1s - loss: 8.8819e-04 - acc: 0.8347 - val_loss: 0.0015 - val_acc: 0.7313 Epoch 1584/3000 1712/1712 [==============================] - 1s - loss: 7.3157e-04 - acc: 0.8300 - val_loss: 0.0020 - val_acc: 0.7383 Epoch 1585/3000 1712/1712 [==============================] - 1s - loss: 8.5522e-04 - acc: 0.8306 - val_loss: 0.0020 - val_acc: 0.7407 Epoch 1586/3000 1712/1712 [==============================] - 1s - loss: 8.1203e-04 - acc: 0.8388 - val_loss: 0.0022 - val_acc: 0.7266 Epoch 1587/3000 1712/1712 [==============================] - 1s - loss: 9.3181e-04 - acc: 0.8306 - val_loss: 0.0017 - val_acc: 0.7640 Epoch 1588/3000 1712/1712 [==============================] - 1s - loss: 7.3390e-04 - acc: 0.8376 - val_loss: 0.0014 - val_acc: 0.7757 Epoch 1589/3000 1712/1712 [==============================] - 1s - loss: 8.9074e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.8107 Epoch 1590/3000 1712/1712 [==============================] - 1s - loss: 8.0453e-04 - acc: 0.8289 - val_loss: 0.0017 - val_acc: 0.7570 Epoch 1591/3000 1712/1712 [==============================] - 1s - loss: 8.2583e-04 - acc: 0.8113 - val_loss: 0.0016 - val_acc: 0.7477 Epoch 1592/3000 1712/1712 [==============================] - 1s - loss: 7.7162e-04 - acc: 0.8312 - val_loss: 0.0020 - val_acc: 0.7290 Epoch 1593/3000 1712/1712 [==============================] - 1s - loss: 7.7309e-04 - acc: 0.8411 - val_loss: 0.0022 - val_acc: 0.7383 Epoch 1594/3000 1712/1712 [==============================] - 1s - loss: 8.4526e-04 - acc: 0.8236 - val_loss: 0.0020 - val_acc: 0.7593 Epoch 1595/3000 1712/1712 [==============================] - 1s - loss: 7.9938e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1596/3000 1712/1712 [==============================] - 1s - loss: 7.9341e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1597/3000 1712/1712 [==============================] - 1s - loss: 8.0368e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 1598/3000 1712/1712 [==============================] - 1s - loss: 7.8132e-04 - acc: 0.8405 - val_loss: 0.0015 - val_acc: 0.7967 Epoch 1599/3000 1712/1712 [==============================] - 1s - loss: 8.2027e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 1600/3000 1712/1712 [==============================] - 1s - loss: 7.9802e-04 - acc: 0.8511 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1601/3000 1712/1712 [==============================] - 1s - loss: 7.4166e-04 - acc: 0.8499 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1602/3000 1712/1712 [==============================] - 1s - loss: 8.5479e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7547 Epoch 1603/3000 1712/1712 [==============================] - 1s - loss: 7.5566e-04 - acc: 0.8242 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1604/3000 1712/1712 [==============================] - 1s - loss: 8.3297e-04 - acc: 0.8458 - val_loss: 0.0013 - val_acc: 0.8201 Epoch 1605/3000 1712/1712 [==============================] - 1s - loss: 7.8409e-04 - acc: 0.8452 - val_loss: 0.0013 - val_acc: 0.7500 Epoch 1606/3000 1712/1712 [==============================] - 1s - loss: 8.4439e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.7500 Epoch 1607/3000 1712/1712 [==============================] - 1s - loss: 7.7585e-04 - acc: 0.8335 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 1608/3000 1712/1712 [==============================] - 1s - loss: 8.2984e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 1609/3000 1712/1712 [==============================] - 1s - loss: 7.9561e-04 - acc: 0.8364 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1610/3000 1712/1712 [==============================] - 1s - loss: 7.9524e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 1611/3000 1712/1712 [==============================] - 1s - loss: 7.8707e-04 - acc: 0.8300 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 1612/3000 1712/1712 [==============================] - 1s - loss: 8.6017e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 1613/3000 1712/1712 [==============================] - 1s - loss: 7.8858e-04 - acc: 0.8300 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1614/3000 1712/1712 [==============================] - 1s - loss: 7.7833e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 1615/3000 1712/1712 [==============================] - 1s - loss: 8.3078e-04 - acc: 0.8400 - val_loss: 0.0015 - val_acc: 0.7593 Epoch 1616/3000 1712/1712 [==============================] - 1s - loss: 9.4015e-04 - acc: 0.8283 - val_loss: 0.0015 - val_acc: 0.7196 Epoch 1617/3000 1712/1712 [==============================] - 1s - loss: 7.4059e-04 - acc: 0.8452 - val_loss: 0.0017 - val_acc: 0.7640 Epoch 1618/3000 1712/1712 [==============================] - 1s - loss: 8.0500e-04 - acc: 0.8359 - val_loss: 0.0018 - val_acc: 0.7079 Epoch 1619/3000 1712/1712 [==============================] - 1s - loss: 8.0939e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1620/3000 1712/1712 [==============================] - 1s - loss: 7.9809e-04 - acc: 0.8277 - val_loss: 0.0017 - val_acc: 0.7500 Epoch 1621/3000 1712/1712 [==============================] - 1s - loss: 8.0737e-04 - acc: 0.8370 - val_loss: 0.0020 - val_acc: 0.7360 Epoch 1622/3000 1712/1712 [==============================] - 1s - loss: 8.4232e-04 - acc: 0.8294 - val_loss: 0.0016 - val_acc: 0.7687 Epoch 1623/3000 1712/1712 [==============================] - 1s - loss: 8.2086e-04 - acc: 0.8359 - val_loss: 0.0028 - val_acc: 0.7009 Epoch 1624/3000 1712/1712 [==============================] - 1s - loss: 8.1444e-04 - acc: 0.8400 - val_loss: 0.0025 - val_acc: 0.7360 Epoch 1625/3000 1712/1712 [==============================] - 1s - loss: 8.2015e-04 - acc: 0.8435 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 1626/3000 1712/1712 [==============================] - 1s - loss: 7.8270e-04 - acc: 0.8411 - val_loss: 0.0011 - val_acc: 0.7921 Epoch 1627/3000 1712/1712 [==============================] - 1s - loss: 8.5644e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.8061 Epoch 1628/3000 1712/1712 [==============================] - 1s - loss: 8.3096e-04 - acc: 0.8318 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1629/3000 1712/1712 [==============================] - 1s - loss: 7.9837e-04 - acc: 0.8324 - val_loss: 0.0016 - val_acc: 0.7827 Epoch 1630/3000 1712/1712 [==============================] - 1s - loss: 8.3488e-04 - acc: 0.8265 - val_loss: 0.0018 - val_acc: 0.7547 Epoch 1631/3000 1712/1712 [==============================] - 1s - loss: 7.3424e-04 - acc: 0.8347 - val_loss: 0.0016 - val_acc: 0.7827 Epoch 1632/3000 1712/1712 [==============================] - 1s - loss: 8.2262e-04 - acc: 0.8417 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1633/3000 1712/1712 [==============================] - 1s - loss: 8.4861e-04 - acc: 0.8300 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1634/3000 1712/1712 [==============================] - 1s - loss: 7.7535e-04 - acc: 0.8254 - val_loss: 0.0021 - val_acc: 0.7243 Epoch 1635/3000 1712/1712 [==============================] - 1s - loss: 7.7048e-04 - acc: 0.8329 - val_loss: 0.0021 - val_acc: 0.7430 Epoch 1636/3000 1712/1712 [==============================] - 1s - loss: 8.1565e-04 - acc: 0.8265 - val_loss: 0.0026 - val_acc: 0.7103 Epoch 1637/3000 1712/1712 [==============================] - 1s - loss: 8.1711e-04 - acc: 0.8289 - val_loss: 0.0014 - val_acc: 0.7874 Epoch 1638/3000 1712/1712 [==============================] - 1s - loss: 7.7464e-04 - acc: 0.8405 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 1639/3000 1712/1712 [==============================] - 1s - loss: 8.5054e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 1640/3000 1712/1712 [==============================] - 1s - loss: 8.4023e-04 - acc: 0.8394 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 1641/3000 1712/1712 [==============================] - 1s - loss: 7.8401e-04 - acc: 0.8353 - val_loss: 0.0017 - val_acc: 0.7336 Epoch 1642/3000 1712/1712 [==============================] - 1s - loss: 8.0048e-04 - acc: 0.8277 - val_loss: 0.0018 - val_acc: 0.7313 Epoch 1643/3000 1712/1712 [==============================] - 1s - loss: 8.3144e-04 - acc: 0.8318 - val_loss: 0.0018 - val_acc: 0.7500 Epoch 1644/3000 1712/1712 [==============================] - 1s - loss: 8.1728e-04 - acc: 0.8318 - val_loss: 0.0016 - val_acc: 0.7570 Epoch 1645/3000 1712/1712 [==============================] - 1s - loss: 8.3794e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 1646/3000 1712/1712 [==============================] - 1s - loss: 7.3509e-04 - acc: 0.8417 - val_loss: 0.0012 - val_acc: 0.8154 Epoch 1647/3000 1712/1712 [==============================] - 1s - loss: 8.0064e-04 - acc: 0.8271 - val_loss: 0.0014 - val_acc: 0.7617 Epoch 1648/3000 1712/1712 [==============================] - 1s - loss: 8.2778e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1649/3000 1712/1712 [==============================] - 1s - loss: 7.7896e-04 - acc: 0.8435 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1650/3000 1712/1712 [==============================] - 1s - loss: 7.9533e-04 - acc: 0.8370 - val_loss: 0.0013 - val_acc: 0.7547 Epoch 1651/3000 1712/1712 [==============================] - 1s - loss: 8.0257e-04 - acc: 0.8464 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1652/3000 1712/1712 [==============================] - 1s - loss: 8.2902e-04 - acc: 0.8376 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 1653/3000 1712/1712 [==============================] - 1s - loss: 7.6420e-04 - acc: 0.8452 - val_loss: 0.0019 - val_acc: 0.7407 Epoch 1654/3000 1712/1712 [==============================] - 1s - loss: 8.0838e-04 - acc: 0.8230 - val_loss: 0.0020 - val_acc: 0.7453 Epoch 1655/3000 1712/1712 [==============================] - 1s - loss: 7.7102e-04 - acc: 0.8359 - val_loss: 0.0017 - val_acc: 0.7336 Epoch 1656/3000 1712/1712 [==============================] - 1s - loss: 8.1372e-04 - acc: 0.8259 - val_loss: 0.0020 - val_acc: 0.7430 Epoch 1657/3000 1712/1712 [==============================] - 1s - loss: 8.6679e-04 - acc: 0.8382 - val_loss: 0.0020 - val_acc: 0.7336 Epoch 1658/3000 1712/1712 [==============================] - 1s - loss: 7.9537e-04 - acc: 0.8312 - val_loss: 0.0020 - val_acc: 0.7313 Epoch 1659/3000 1712/1712 [==============================] - 1s - loss: 8.0144e-04 - acc: 0.8546 - val_loss: 0.0020 - val_acc: 0.7266 Epoch 1660/3000 1712/1712 [==============================] - 1s - loss: 8.2888e-04 - acc: 0.8382 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 1661/3000 1712/1712 [==============================] - 1s - loss: 7.9330e-04 - acc: 0.8335 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 1662/3000 1712/1712 [==============================] - 1s - loss: 8.2943e-04 - acc: 0.8329 - val_loss: 0.0016 - val_acc: 0.7827 Epoch 1663/3000 1712/1712 [==============================] - 1s - loss: 7.5468e-04 - acc: 0.8458 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 1664/3000 1712/1712 [==============================] - 1s - loss: 8.2246e-04 - acc: 0.8283 - val_loss: 0.0014 - val_acc: 0.7734 Epoch 1665/3000 1712/1712 [==============================] - 1s - loss: 7.5127e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1666/3000 1712/1712 [==============================] - 1s - loss: 8.1631e-04 - acc: 0.8242 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 1667/3000 1712/1712 [==============================] - 1s - loss: 8.5988e-04 - acc: 0.8394 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1668/3000 1712/1712 [==============================] - 1s - loss: 8.4034e-04 - acc: 0.8236 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 1669/3000 1712/1712 [==============================] - 1s - loss: 7.8330e-04 - acc: 0.8481 - val_loss: 0.0014 - val_acc: 0.7827 Epoch 1670/3000 1712/1712 [==============================] - 1s - loss: 7.9301e-04 - acc: 0.8329 - val_loss: 0.0014 - val_acc: 0.7617 Epoch 1671/3000 1712/1712 [==============================] - 1s - loss: 8.0995e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1672/3000 1712/1712 [==============================] - 1s - loss: 7.6936e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 1673/3000 1712/1712 [==============================] - 1s - loss: 8.5119e-04 - acc: 0.8160 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1674/3000 1712/1712 [==============================] - 1s - loss: 7.5682e-04 - acc: 0.8353 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 1675/3000 1712/1712 [==============================] - 1s - loss: 7.9049e-04 - acc: 0.8417 - val_loss: 0.0017 - val_acc: 0.7640 Epoch 1676/3000 1712/1712 [==============================] - 1s - loss: 8.6256e-04 - acc: 0.8248 - val_loss: 0.0015 - val_acc: 0.7827 Epoch 1677/3000 1712/1712 [==============================] - 1s - loss: 6.7511e-04 - acc: 0.8435 - val_loss: 0.0019 - val_acc: 0.7103 Epoch 1678/3000 1712/1712 [==============================] - 1s - loss: 8.5695e-04 - acc: 0.8084 - val_loss: 0.0019 - val_acc: 0.7360 Epoch 1679/3000 1712/1712 [==============================] - 1s - loss: 8.4663e-04 - acc: 0.8236 - val_loss: 0.0021 - val_acc: 0.7407 Epoch 1680/3000 1712/1712 [==============================] - 1s - loss: 8.1073e-04 - acc: 0.8417 - val_loss: 0.0019 - val_acc: 0.7243 Epoch 1681/3000 1712/1712 [==============================] - 1s - loss: 8.0126e-04 - acc: 0.8300 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 1682/3000 1712/1712 [==============================] - 1s - loss: 7.8703e-04 - acc: 0.8236 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 1683/3000 1712/1712 [==============================] - 1s - loss: 7.9552e-04 - acc: 0.8254 - val_loss: 0.0019 - val_acc: 0.7336 Epoch 1684/3000 1712/1712 [==============================] - 1s - loss: 8.3416e-04 - acc: 0.8207 - val_loss: 0.0020 - val_acc: 0.7477 Epoch 1685/3000 1712/1712 [==============================] - 1s - loss: 8.2922e-04 - acc: 0.8312 - val_loss: 0.0022 - val_acc: 0.7243 Epoch 1686/3000 1712/1712 [==============================] - 1s - loss: 8.1186e-04 - acc: 0.8405 - val_loss: 0.0015 - val_acc: 0.7360 Epoch 1687/3000 1712/1712 [==============================] - 1s - loss: 8.4536e-04 - acc: 0.8277 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 1688/3000 1712/1712 [==============================] - 1s - loss: 8.4333e-04 - acc: 0.8312 - val_loss: 0.0019 - val_acc: 0.7360 Epoch 1689/3000 1712/1712 [==============================] - 1s - loss: 7.5713e-04 - acc: 0.8452 - val_loss: 0.0020 - val_acc: 0.7243 Epoch 1690/3000 1712/1712 [==============================] - 1s - loss: 9.0619e-04 - acc: 0.8254 - val_loss: 0.0014 - val_acc: 0.8014 Epoch 1691/3000 1712/1712 [==============================] - 1s - loss: 7.1753e-04 - acc: 0.8283 - val_loss: 0.0014 - val_acc: 0.7407 Epoch 1692/3000 1712/1712 [==============================] - 1s - loss: 8.0563e-04 - acc: 0.8230 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 1693/3000 1712/1712 [==============================] - 1s - loss: 7.9422e-04 - acc: 0.8376 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 1694/3000 1712/1712 [==============================] - 1s - loss: 7.5733e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 1695/3000 1712/1712 [==============================] - 1s - loss: 8.2550e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 1696/3000 1712/1712 [==============================] - 1s - loss: 8.4058e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1697/3000 1712/1712 [==============================] - 1s - loss: 7.9896e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1698/3000 1712/1712 [==============================] - 1s - loss: 8.4937e-04 - acc: 0.8259 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1699/3000 1712/1712 [==============================] - 1s - loss: 7.7777e-04 - acc: 0.8224 - val_loss: 0.0013 - val_acc: 0.8107 Epoch 1700/3000 1712/1712 [==============================] - 1s - loss: 8.1521e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 1701/3000 1712/1712 [==============================] - 1s - loss: 7.8159e-04 - acc: 0.8148 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 1702/3000 1712/1712 [==============================] - 1s - loss: 8.2154e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 1703/3000 1712/1712 [==============================] - 1s - loss: 8.2334e-04 - acc: 0.8277 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 1704/3000 1712/1712 [==============================] - 1s - loss: 8.0529e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1705/3000 1712/1712 [==============================] - 1s - loss: 8.5643e-04 - acc: 0.8400 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 1706/3000 1712/1712 [==============================] - 1s - loss: 7.3202e-04 - acc: 0.8470 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 1707/3000 1712/1712 [==============================] - 1s - loss: 8.5328e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 1708/3000 1712/1712 [==============================] - 1s - loss: 7.9214e-04 - acc: 0.8300 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 1709/3000 1712/1712 [==============================] - 1s - loss: 7.9399e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7617 Epoch 1710/3000 1712/1712 [==============================] - 1s - loss: 7.5842e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.8224 Epoch 1711/3000 1712/1712 [==============================] - 1s - loss: 8.8502e-04 - acc: 0.8370 - val_loss: 0.0014 - val_acc: 0.7664 Epoch 1712/3000 1712/1712 [==============================] - 1s - loss: 7.6166e-04 - acc: 0.8394 - val_loss: 0.0015 - val_acc: 0.7430 Epoch 1713/3000 1712/1712 [==============================] - 1s - loss: 7.3893e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.7640 Epoch 1714/3000 1712/1712 [==============================] - 1s - loss: 8.2362e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1715/3000 1712/1712 [==============================] - 1s - loss: 8.3778e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 1716/3000 1712/1712 [==============================] - 1s - loss: 8.1217e-04 - acc: 0.8382 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 1717/3000 1712/1712 [==============================] - 1s - loss: 7.6963e-04 - acc: 0.8458 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 1718/3000 1712/1712 [==============================] - 1s - loss: 8.9746e-04 - acc: 0.8178 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 1719/3000 1712/1712 [==============================] - 1s - loss: 7.9411e-04 - acc: 0.8353 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1720/3000 1712/1712 [==============================] - 1s - loss: 8.1649e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 1721/3000 1712/1712 [==============================] - 1s - loss: 8.3953e-04 - acc: 0.8470 - val_loss: 0.0012 - val_acc: 0.7500 Epoch 1722/3000 1712/1712 [==============================] - 1s - loss: 7.8279e-04 - acc: 0.8364 - val_loss: 0.0011 - val_acc: 0.7967 Epoch 1723/3000 1712/1712 [==============================] - 1s - loss: 7.9491e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 1724/3000 1712/1712 [==============================] - 1s - loss: 7.9436e-04 - acc: 0.8400 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1725/3000 1712/1712 [==============================] - 1s - loss: 8.0565e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1726/3000 1712/1712 [==============================] - 1s - loss: 8.4619e-04 - acc: 0.8318 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 1727/3000 1712/1712 [==============================] - 1s - loss: 7.8680e-04 - acc: 0.8511 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 1728/3000 1712/1712 [==============================] - 1s - loss: 7.8304e-04 - acc: 0.8230 - val_loss: 0.0020 - val_acc: 0.7173 Epoch 1729/3000 1712/1712 [==============================] - 1s - loss: 8.9934e-04 - acc: 0.8148 - val_loss: 0.0018 - val_acc: 0.7570 Epoch 1730/3000 1712/1712 [==============================] - 1s - loss: 7.2833e-04 - acc: 0.8364 - val_loss: 0.0017 - val_acc: 0.7313 Epoch 1731/3000 1712/1712 [==============================] - 1s - loss: 8.4302e-04 - acc: 0.8248 - val_loss: 0.0019 - val_acc: 0.7430 Epoch 1732/3000 1712/1712 [==============================] - 1s - loss: 7.9013e-04 - acc: 0.8388 - val_loss: 0.0020 - val_acc: 0.7547 Epoch 1733/3000 1712/1712 [==============================] - 1s - loss: 8.2334e-04 - acc: 0.8417 - val_loss: 0.0017 - val_acc: 0.7360 Epoch 1734/3000 1712/1712 [==============================] - 1s - loss: 8.7606e-04 - acc: 0.8218 - val_loss: 0.0019 - val_acc: 0.7126 Epoch 1735/3000 1712/1712 [==============================] - 1s - loss: 8.0394e-04 - acc: 0.8359 - val_loss: 0.0017 - val_acc: 0.7453 Epoch 1736/3000 1712/1712 [==============================] - 1s - loss: 8.1824e-04 - acc: 0.8324 - val_loss: 0.0015 - val_acc: 0.7570 Epoch 1737/3000 1712/1712 [==============================] - 1s - loss: 8.1579e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 1738/3000 1712/1712 [==============================] - 1s - loss: 7.7936e-04 - acc: 0.8394 - val_loss: 0.0014 - val_acc: 0.7640 Epoch 1739/3000 1712/1712 [==============================] - 1s - loss: 8.3029e-04 - acc: 0.8259 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 1740/3000 1712/1712 [==============================] - 1s - loss: 7.9036e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 1741/3000 1712/1712 [==============================] - 1s - loss: 8.1845e-04 - acc: 0.8394 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1742/3000 1712/1712 [==============================] - 1s - loss: 7.6450e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1743/3000 1712/1712 [==============================] - 1s - loss: 7.8633e-04 - acc: 0.8300 - val_loss: 0.0014 - val_acc: 0.7757 Epoch 1744/3000 1712/1712 [==============================] - 1s - loss: 8.4617e-04 - acc: 0.8353 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 1745/3000 1712/1712 [==============================] - 1s - loss: 7.7327e-04 - acc: 0.8329 - val_loss: 0.0020 - val_acc: 0.7570 Epoch 1746/3000 1712/1712 [==============================] - 1s - loss: 8.2269e-04 - acc: 0.8388 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 1747/3000 1712/1712 [==============================] - 1s - loss: 7.6273e-04 - acc: 0.8353 - val_loss: 0.0021 - val_acc: 0.7196 Epoch 1748/3000 1712/1712 [==============================] - 1s - loss: 8.4588e-04 - acc: 0.8318 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 1749/3000 1712/1712 [==============================] - 1s - loss: 7.5706e-04 - acc: 0.8324 - val_loss: 0.0018 - val_acc: 0.7710 Epoch 1750/3000 1712/1712 [==============================] - 1s - loss: 7.9727e-04 - acc: 0.8347 - val_loss: 0.0017 - val_acc: 0.7290 Epoch 1751/3000 1712/1712 [==============================] - 1s - loss: 8.2193e-04 - acc: 0.8277 - val_loss: 0.0020 - val_acc: 0.7290 Epoch 1752/3000 1712/1712 [==============================] - 1s - loss: 8.4211e-04 - acc: 0.8405 - val_loss: 0.0021 - val_acc: 0.7360 Epoch 1753/3000 1712/1712 [==============================] - 1s - loss: 8.0457e-04 - acc: 0.8347 - val_loss: 0.0017 - val_acc: 0.7243 Epoch 1754/3000 1712/1712 [==============================] - 1s - loss: 8.6219e-04 - acc: 0.8283 - val_loss: 0.0020 - val_acc: 0.7500 Epoch 1755/3000 1712/1712 [==============================] - 1s - loss: 7.8299e-04 - acc: 0.8405 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 1756/3000 1712/1712 [==============================] - 1s - loss: 7.9486e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 1757/3000 1712/1712 [==============================] - 1s - loss: 8.1621e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 1758/3000 1712/1712 [==============================] - 1s - loss: 8.1603e-04 - acc: 0.8516 - val_loss: 0.0019 - val_acc: 0.7477 Epoch 1759/3000 1712/1712 [==============================] - 1s - loss: 8.0058e-04 - acc: 0.8370 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 1760/3000 1712/1712 [==============================] - 1s - loss: 8.2166e-04 - acc: 0.8224 - val_loss: 0.0024 - val_acc: 0.7780 Epoch 1761/3000 1712/1712 [==============================] - 1s - loss: 8.2887e-04 - acc: 0.8364 - val_loss: 0.0017 - val_acc: 0.7383 Epoch 1762/3000 1712/1712 [==============================] - 1s - loss: 7.6480e-04 - acc: 0.8429 - val_loss: 0.0016 - val_acc: 0.7547 Epoch 1763/3000 1712/1712 [==============================] - 1s - loss: 7.7944e-04 - acc: 0.8254 - val_loss: 0.0018 - val_acc: 0.7196 Epoch 1764/3000 1712/1712 [==============================] - 1s - loss: 8.3898e-04 - acc: 0.8201 - val_loss: 0.0015 - val_acc: 0.7593 Epoch 1765/3000 1712/1712 [==============================] - 1s - loss: 8.0409e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7664 Epoch 1766/3000 1712/1712 [==============================] - 1s - loss: 7.7286e-04 - acc: 0.8470 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 1767/3000 1712/1712 [==============================] - 1s - loss: 7.9292e-04 - acc: 0.8353 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1768/3000 1712/1712 [==============================] - 1s - loss: 8.5759e-04 - acc: 0.8213 - val_loss: 0.0011 - val_acc: 0.8294 Epoch 1769/3000 1712/1712 [==============================] - 1s - loss: 7.4478e-04 - acc: 0.8429 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1770/3000 1712/1712 [==============================] - 1s - loss: 8.9212e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7640 Epoch 1771/3000 1712/1712 [==============================] - 1s - loss: 8.5628e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.7570 Epoch 1772/3000 1712/1712 [==============================] - 1s - loss: 7.6551e-04 - acc: 0.8423 - val_loss: 0.0021 - val_acc: 0.7009 Epoch 1773/3000 1712/1712 [==============================] - 1s - loss: 8.4892e-04 - acc: 0.8394 - val_loss: 0.0015 - val_acc: 0.7570 Epoch 1774/3000 1712/1712 [==============================] - 1s - loss: 9.0557e-04 - acc: 0.8341 - val_loss: 0.0016 - val_acc: 0.7430 Epoch 1775/3000 1712/1712 [==============================] - 1s - loss: 7.6053e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 1776/3000 1712/1712 [==============================] - 1s - loss: 7.7174e-04 - acc: 0.8341 - val_loss: 0.0015 - val_acc: 0.8084 Epoch 1777/3000 1712/1712 [==============================] - 1s - loss: 7.5170e-04 - acc: 0.8481 - val_loss: 0.0024 - val_acc: 0.6986 Epoch 1778/3000 1712/1712 [==============================] - 1s - loss: 7.9130e-04 - acc: 0.8271 - val_loss: 0.0017 - val_acc: 0.7500 Epoch 1779/3000 1712/1712 [==============================] - 1s - loss: 8.9490e-04 - acc: 0.8201 - val_loss: 0.0018 - val_acc: 0.7453 Epoch 1780/3000 1712/1712 [==============================] - 1s - loss: 7.5717e-04 - acc: 0.8306 - val_loss: 0.0016 - val_acc: 0.7710 Epoch 1781/3000 1712/1712 [==============================] - 1s - loss: 8.0374e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 1782/3000 1712/1712 [==============================] - 1s - loss: 7.5206e-04 - acc: 0.8411 - val_loss: 0.0012 - val_acc: 0.7640 Epoch 1783/3000 1712/1712 [==============================] - 1s - loss: 8.3522e-04 - acc: 0.8429 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1784/3000 1712/1712 [==============================] - 1s - loss: 7.9481e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 1785/3000 1712/1712 [==============================] - 1s - loss: 7.5222e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1786/3000 1712/1712 [==============================] - 1s - loss: 8.2688e-04 - acc: 0.8376 - val_loss: 0.0020 - val_acc: 0.7921 Epoch 1787/3000 1712/1712 [==============================] - 1s - loss: 8.2411e-04 - acc: 0.8224 - val_loss: 0.0017 - val_acc: 0.7617 Epoch 1788/3000 1712/1712 [==============================] - 1s - loss: 7.7581e-04 - acc: 0.8446 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 1789/3000 1712/1712 [==============================] - 1s - loss: 8.1296e-04 - acc: 0.8271 - val_loss: 0.0022 - val_acc: 0.7173 Epoch 1790/3000 1712/1712 [==============================] - 1s - loss: 8.1386e-04 - acc: 0.8405 - val_loss: 0.0020 - val_acc: 0.7547 Epoch 1791/3000 1712/1712 [==============================] - 1s - loss: 7.7719e-04 - acc: 0.8271 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 1792/3000 1712/1712 [==============================] - 1s - loss: 9.0310e-04 - acc: 0.8411 - val_loss: 0.0016 - val_acc: 0.7640 Epoch 1793/3000 1712/1712 [==============================] - 1s - loss: 8.4203e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1794/3000 1712/1712 [==============================] - 1s - loss: 7.1817e-04 - acc: 0.8400 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 1795/3000 1712/1712 [==============================] - 1s - loss: 8.3677e-04 - acc: 0.8353 - val_loss: 0.0013 - val_acc: 0.7547 Epoch 1796/3000 1712/1712 [==============================] - 1s - loss: 7.7972e-04 - acc: 0.8370 - val_loss: 0.0014 - val_acc: 0.7430 Epoch 1797/3000 1712/1712 [==============================] - 1s - loss: 8.0105e-04 - acc: 0.8440 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 1798/3000 1712/1712 [==============================] - 1s - loss: 8.2414e-04 - acc: 0.8516 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1799/3000 1712/1712 [==============================] - 1s - loss: 8.1817e-04 - acc: 0.8300 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1800/3000 1712/1712 [==============================] - 1s - loss: 7.7305e-04 - acc: 0.8429 - val_loss: 0.0019 - val_acc: 0.7617 Epoch 1801/3000 1712/1712 [==============================] - 1s - loss: 8.6641e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 1802/3000 1712/1712 [==============================] - 1s - loss: 7.6643e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 1803/3000 1712/1712 [==============================] - 1s - loss: 8.3183e-04 - acc: 0.8242 - val_loss: 0.0014 - val_acc: 0.7477 Epoch 1804/3000 1712/1712 [==============================] - 1s - loss: 8.0056e-04 - acc: 0.8400 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 1805/3000 1712/1712 [==============================] - 1s - loss: 8.0828e-04 - acc: 0.8470 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 1806/3000 1712/1712 [==============================] - 1s - loss: 8.1271e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 1807/3000 1712/1712 [==============================] - 1s - loss: 7.5682e-04 - acc: 0.8242 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 1808/3000 1712/1712 [==============================] - 1s - loss: 7.9549e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 1809/3000 1712/1712 [==============================] - 1s - loss: 7.8214e-04 - acc: 0.8405 - val_loss: 0.0017 - val_acc: 0.7336 Epoch 1810/3000 1712/1712 [==============================] - 1s - loss: 8.1540e-04 - acc: 0.8294 - val_loss: 0.0020 - val_acc: 0.7126 Epoch 1811/3000 1712/1712 [==============================] - 1s - loss: 8.2729e-04 - acc: 0.8242 - val_loss: 0.0017 - val_acc: 0.7313 Epoch 1812/3000 1712/1712 [==============================] - 1s - loss: 7.4255e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7664 Epoch 1813/3000 1712/1712 [==============================] - 1s - loss: 9.7355e-04 - acc: 0.8178 - val_loss: 0.0016 - val_acc: 0.7430 Epoch 1814/3000 1712/1712 [==============================] - 1s - loss: 7.5953e-04 - acc: 0.8347 - val_loss: 0.0015 - val_acc: 0.7243 Epoch 1815/3000 1712/1712 [==============================] - 1s - loss: 8.0651e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 1816/3000 1712/1712 [==============================] - 1s - loss: 7.7311e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 1817/3000 1712/1712 [==============================] - 1s - loss: 7.9089e-04 - acc: 0.8400 - val_loss: 0.0014 - val_acc: 0.7734 Epoch 1818/3000 1712/1712 [==============================] - 1s - loss: 8.2534e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 1819/3000 1712/1712 [==============================] - 1s - loss: 7.6915e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 1820/3000 1712/1712 [==============================] - 1s - loss: 7.6993e-04 - acc: 0.8481 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 1821/3000 1712/1712 [==============================] - 1s - loss: 7.7998e-04 - acc: 0.8277 - val_loss: 0.0021 - val_acc: 0.7220 Epoch 1822/3000 1712/1712 [==============================] - 1s - loss: 7.9735e-04 - acc: 0.8376 - val_loss: 0.0022 - val_acc: 0.7150 Epoch 1823/3000 1712/1712 [==============================] - 1s - loss: 8.4714e-04 - acc: 0.8265 - val_loss: 0.0016 - val_acc: 0.7780 Epoch 1824/3000 1712/1712 [==============================] - 1s - loss: 8.0919e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 1825/3000 1712/1712 [==============================] - 1s - loss: 7.6800e-04 - acc: 0.8551 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 1826/3000 1712/1712 [==============================] - 1s - loss: 8.1731e-04 - acc: 0.8470 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 1827/3000 1712/1712 [==============================] - 1s - loss: 7.9619e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7593 Epoch 1828/3000 1712/1712 [==============================] - 1s - loss: 7.3209e-04 - acc: 0.8411 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 1829/3000 1712/1712 [==============================] - 1s - loss: 8.5768e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7523 Epoch 1830/3000 1712/1712 [==============================] - 1s - loss: 7.7081e-04 - acc: 0.8382 - val_loss: 0.0012 - val_acc: 0.7547 Epoch 1831/3000 1712/1712 [==============================] - 1s - loss: 8.1828e-04 - acc: 0.8230 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 1832/3000 1712/1712 [==============================] - 1s - loss: 7.9446e-04 - acc: 0.8259 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 1833/3000 1712/1712 [==============================] - 1s - loss: 7.9125e-04 - acc: 0.8417 - val_loss: 0.0024 - val_acc: 0.7360 Epoch 1834/3000 1712/1712 [==============================] - 1s - loss: 8.0773e-04 - acc: 0.8306 - val_loss: 0.0020 - val_acc: 0.7196 Epoch 1835/3000 1712/1712 [==============================] - 1s - loss: 7.9035e-04 - acc: 0.8341 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 1836/3000 1712/1712 [==============================] - 1s - loss: 8.2210e-04 - acc: 0.8236 - val_loss: 0.0021 - val_acc: 0.7336 Epoch 1837/3000 1712/1712 [==============================] - 1s - loss: 8.4077e-04 - acc: 0.8195 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 1838/3000 1712/1712 [==============================] - 1s - loss: 7.8234e-04 - acc: 0.8166 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 1839/3000 1712/1712 [==============================] - 1s - loss: 8.1993e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 1840/3000 1712/1712 [==============================] - 1s - loss: 7.7326e-04 - acc: 0.8242 - val_loss: 0.0014 - val_acc: 0.7593 Epoch 1841/3000 1712/1712 [==============================] - 1s - loss: 7.8916e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7547 Epoch 1842/3000 1712/1712 [==============================] - 1s - loss: 8.3457e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 1843/3000 1712/1712 [==============================] - 1s - loss: 7.9682e-04 - acc: 0.8423 - val_loss: 0.0013 - val_acc: 0.8154 Epoch 1844/3000 1712/1712 [==============================] - 1s - loss: 7.7647e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1845/3000 1712/1712 [==============================] - 1s - loss: 7.5410e-04 - acc: 0.8417 - val_loss: 0.0019 - val_acc: 0.7336 Epoch 1846/3000 1712/1712 [==============================] - 1s - loss: 8.2030e-04 - acc: 0.8388 - val_loss: 0.0021 - val_acc: 0.7196 Epoch 1847/3000 1712/1712 [==============================] - 1s - loss: 8.3686e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 1848/3000 1712/1712 [==============================] - 1s - loss: 8.0210e-04 - acc: 0.8405 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 1849/3000 1712/1712 [==============================] - 1s - loss: 7.9098e-04 - acc: 0.8382 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 1850/3000 1712/1712 [==============================] - 1s - loss: 8.6851e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 1851/3000 1712/1712 [==============================] - 1s - loss: 7.9007e-04 - acc: 0.8230 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 1852/3000 1712/1712 [==============================] - 1s - loss: 7.8091e-04 - acc: 0.8411 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1853/3000 1712/1712 [==============================] - 1s - loss: 8.2496e-04 - acc: 0.8300 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1854/3000 1712/1712 [==============================] - 1s - loss: 8.0425e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7640 Epoch 1855/3000 1712/1712 [==============================] - 1s - loss: 7.8479e-04 - acc: 0.8353 - val_loss: 0.0017 - val_acc: 0.7757 Epoch 1856/3000 1712/1712 [==============================] - 1s - loss: 7.6120e-04 - acc: 0.8353 - val_loss: 0.0024 - val_acc: 0.7313 Epoch 1857/3000 1712/1712 [==============================] - 1s - loss: 8.6312e-04 - acc: 0.8207 - val_loss: 0.0019 - val_acc: 0.7290 Epoch 1858/3000 1712/1712 [==============================] - 1s - loss: 8.2738e-04 - acc: 0.8306 - val_loss: 0.0018 - val_acc: 0.7640 Epoch 1859/3000 1712/1712 [==============================] - 1s - loss: 8.2649e-04 - acc: 0.8294 - val_loss: 0.0018 - val_acc: 0.7617 Epoch 1860/3000 1712/1712 [==============================] - 1s - loss: 7.6096e-04 - acc: 0.8440 - val_loss: 0.0017 - val_acc: 0.7710 Epoch 1861/3000 1712/1712 [==============================] - 1s - loss: 8.1423e-04 - acc: 0.8376 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 1862/3000 1712/1712 [==============================] - 1s - loss: 7.9806e-04 - acc: 0.8423 - val_loss: 0.0013 - val_acc: 0.8131 Epoch 1863/3000 1712/1712 [==============================] - 1s - loss: 8.4076e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 1864/3000 1712/1712 [==============================] - 1s - loss: 8.4287e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1865/3000 1712/1712 [==============================] - 1s - loss: 7.3993e-04 - acc: 0.8534 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 1866/3000 1712/1712 [==============================] - 1s - loss: 8.0702e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 1867/3000 1712/1712 [==============================] - 1s - loss: 7.9221e-04 - acc: 0.8329 - val_loss: 0.0022 - val_acc: 0.7313 Epoch 1868/3000 1712/1712 [==============================] - 1s - loss: 8.0997e-04 - acc: 0.8411 - val_loss: 0.0019 - val_acc: 0.7266 Epoch 1869/3000 1712/1712 [==============================] - 1s - loss: 7.8613e-04 - acc: 0.8265 - val_loss: 0.0023 - val_acc: 0.7056 Epoch 1870/3000 1712/1712 [==============================] - 1s - loss: 8.2576e-04 - acc: 0.8224 - val_loss: 0.0012 - val_acc: 0.7617 Epoch 1871/3000 1712/1712 [==============================] - 1s - loss: 7.6261e-04 - acc: 0.8417 - val_loss: 0.0018 - val_acc: 0.7383 Epoch 1872/3000 1712/1712 [==============================] - 1s - loss: 8.8161e-04 - acc: 0.8370 - val_loss: 0.0016 - val_acc: 0.8084 Epoch 1873/3000 1712/1712 [==============================] - 1s - loss: 7.3725e-04 - acc: 0.8277 - val_loss: 0.0023 - val_acc: 0.7220 Epoch 1874/3000 1712/1712 [==============================] - 1s - loss: 8.4577e-04 - acc: 0.8242 - val_loss: 0.0022 - val_acc: 0.7220 Epoch 1875/3000 1712/1712 [==============================] - 1s - loss: 7.9450e-04 - acc: 0.8347 - val_loss: 0.0019 - val_acc: 0.7547 Epoch 1876/3000 1712/1712 [==============================] - 1s - loss: 8.2486e-04 - acc: 0.8423 - val_loss: 0.0020 - val_acc: 0.7313 Epoch 1877/3000 1712/1712 [==============================] - 1s - loss: 8.1329e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7453 Epoch 1878/3000 1712/1712 [==============================] - 1s - loss: 7.5731e-04 - acc: 0.8335 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 1879/3000 1712/1712 [==============================] - 1s - loss: 8.5429e-04 - acc: 0.8417 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1880/3000 1712/1712 [==============================] - 1s - loss: 8.4102e-04 - acc: 0.8277 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 1881/3000 1712/1712 [==============================] - 1s - loss: 7.9115e-04 - acc: 0.8440 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 1882/3000 1712/1712 [==============================] - 1s - loss: 7.7084e-04 - acc: 0.8411 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 1883/3000 1712/1712 [==============================] - 1s - loss: 8.9548e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1884/3000 1712/1712 [==============================] - 1s - loss: 7.5934e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.7640 Epoch 1885/3000 1712/1712 [==============================] - 1s - loss: 7.7322e-04 - acc: 0.8429 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 1886/3000 1712/1712 [==============================] - 1s - loss: 8.8750e-04 - acc: 0.8201 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 1887/3000 1712/1712 [==============================] - 1s - loss: 7.4716e-04 - acc: 0.8493 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 1888/3000 1712/1712 [==============================] - 1s - loss: 8.5566e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 1889/3000 1712/1712 [==============================] - 1s - loss: 8.2348e-04 - acc: 0.8195 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 1890/3000 1712/1712 [==============================] - 1s - loss: 8.3243e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 1891/3000 1712/1712 [==============================] - 1s - loss: 8.8492e-04 - acc: 0.8353 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1892/3000 1712/1712 [==============================] - 1s - loss: 7.2024e-04 - acc: 0.8230 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1893/3000 1712/1712 [==============================] - 1s - loss: 8.8218e-04 - acc: 0.8382 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 1894/3000 1712/1712 [==============================] - 1s - loss: 8.0702e-04 - acc: 0.8405 - val_loss: 0.0018 - val_acc: 0.7313 Epoch 1895/3000 1712/1712 [==============================] - 1s - loss: 7.7071e-04 - acc: 0.8347 - val_loss: 0.0023 - val_acc: 0.7220 Epoch 1896/3000 1712/1712 [==============================] - 1s - loss: 8.7400e-04 - acc: 0.8178 - val_loss: 0.0024 - val_acc: 0.7360 Epoch 1897/3000 1712/1712 [==============================] - 1s - loss: 8.0919e-04 - acc: 0.8067 - val_loss: 0.0017 - val_acc: 0.7874 Epoch 1898/3000 1712/1712 [==============================] - 1s - loss: 8.4965e-04 - acc: 0.8218 - val_loss: 0.0017 - val_acc: 0.7687 Epoch 1899/3000 1712/1712 [==============================] - 1s - loss: 7.6173e-04 - acc: 0.8359 - val_loss: 0.0011 - val_acc: 0.7827 Epoch 1900/3000 1712/1712 [==============================] - 1s - loss: 7.8031e-04 - acc: 0.8347 - val_loss: 0.0014 - val_acc: 0.7944 Epoch 1901/3000 1712/1712 [==============================] - 1s - loss: 8.1886e-04 - acc: 0.8306 - val_loss: 0.0017 - val_acc: 0.7570 Epoch 1902/3000 1712/1712 [==============================] - 1s - loss: 7.7648e-04 - acc: 0.8388 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 1903/3000 1712/1712 [==============================] - 1s - loss: 8.5315e-04 - acc: 0.8353 - val_loss: 0.0016 - val_acc: 0.7897 Epoch 1904/3000 1712/1712 [==============================] - 1s - loss: 7.3158e-04 - acc: 0.8335 - val_loss: 0.0025 - val_acc: 0.7056 Epoch 1905/3000 1712/1712 [==============================] - 1s - loss: 8.2949e-04 - acc: 0.8289 - val_loss: 0.0021 - val_acc: 0.7313 Epoch 1906/3000 1712/1712 [==============================] - 1s - loss: 8.1198e-04 - acc: 0.8394 - val_loss: 0.0023 - val_acc: 0.7126 Epoch 1907/3000 1712/1712 [==============================] - 1s - loss: 7.8493e-04 - acc: 0.8335 - val_loss: 0.0022 - val_acc: 0.7173 Epoch 1908/3000 1712/1712 [==============================] - 1s - loss: 8.0351e-04 - acc: 0.8143 - val_loss: 0.0015 - val_acc: 0.7944 Epoch 1909/3000 1712/1712 [==============================] - 1s - loss: 8.2347e-04 - acc: 0.8271 - val_loss: 0.0022 - val_acc: 0.7360 Epoch 1910/3000 1712/1712 [==============================] - 1s - loss: 8.3443e-04 - acc: 0.8364 - val_loss: 0.0017 - val_acc: 0.7687 Epoch 1911/3000 1712/1712 [==============================] - 1s - loss: 7.9287e-04 - acc: 0.8289 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 1912/3000 1712/1712 [==============================] - 1s - loss: 8.3325e-04 - acc: 0.8148 - val_loss: 0.0016 - val_acc: 0.7640 Epoch 1913/3000 1712/1712 [==============================] - 1s - loss: 8.1638e-04 - acc: 0.8400 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 1914/3000 1712/1712 [==============================] - 1s - loss: 7.7980e-04 - acc: 0.8353 - val_loss: 0.0018 - val_acc: 0.7126 Epoch 1915/3000 1712/1712 [==============================] - 1s - loss: 8.2492e-04 - acc: 0.8213 - val_loss: 0.0022 - val_acc: 0.7243 Epoch 1916/3000 1712/1712 [==============================] - 1s - loss: 8.0534e-04 - acc: 0.8289 - val_loss: 0.0018 - val_acc: 0.7710 Epoch 1917/3000 1712/1712 [==============================] - 1s - loss: 8.1312e-04 - acc: 0.8329 - val_loss: 0.0014 - val_acc: 0.7570 Epoch 1918/3000 1712/1712 [==============================] - 1s - loss: 7.9278e-04 - acc: 0.8265 - val_loss: 0.0014 - val_acc: 0.7593 Epoch 1919/3000 1712/1712 [==============================] - 1s - loss: 7.4536e-04 - acc: 0.8481 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 1920/3000 1712/1712 [==============================] - 1s - loss: 8.6462e-04 - acc: 0.8329 - val_loss: 0.0020 - val_acc: 0.7407 Epoch 1921/3000 1712/1712 [==============================] - 1s - loss: 8.3654e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7500 Epoch 1922/3000 1712/1712 [==============================] - 1s - loss: 8.5259e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7547 Epoch 1923/3000 1712/1712 [==============================] - 1s - loss: 7.7307e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 1924/3000 1712/1712 [==============================] - 1s - loss: 9.5339e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7593 Epoch 1925/3000 1712/1712 [==============================] - 1s - loss: 7.3286e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 1926/3000 1712/1712 [==============================] - 1s - loss: 7.8945e-04 - acc: 0.8254 - val_loss: 0.0016 - val_acc: 0.7664 Epoch 1927/3000 1712/1712 [==============================] - 1s - loss: 8.2673e-04 - acc: 0.8289 - val_loss: 0.0023 - val_acc: 0.7033 Epoch 1928/3000 1712/1712 [==============================] - 1s - loss: 8.7695e-04 - acc: 0.8189 - val_loss: 0.0017 - val_acc: 0.7757 Epoch 1929/3000 1712/1712 [==============================] - 1s - loss: 8.0640e-04 - acc: 0.8446 - val_loss: 0.0028 - val_acc: 0.7009 Epoch 1930/3000 1712/1712 [==============================] - 1s - loss: 8.1705e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 1931/3000 1712/1712 [==============================] - 1s - loss: 8.5288e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7593 Epoch 1932/3000 1712/1712 [==============================] - 1s - loss: 8.3732e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 1933/3000 1712/1712 [==============================] - 1s - loss: 7.9666e-04 - acc: 0.8300 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 1934/3000 1712/1712 [==============================] - 1s - loss: 8.1043e-04 - acc: 0.8405 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1935/3000 1712/1712 [==============================] - 1s - loss: 8.2756e-04 - acc: 0.8335 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 1936/3000 1712/1712 [==============================] - 1s - loss: 7.0520e-04 - acc: 0.8470 - val_loss: 0.0025 - val_acc: 0.7173 Epoch 1937/3000 1712/1712 [==============================] - 1s - loss: 8.8539e-04 - acc: 0.8189 - val_loss: 0.0017 - val_acc: 0.7710 Epoch 1938/3000 1712/1712 [==============================] - 1s - loss: 7.4541e-04 - acc: 0.8417 - val_loss: 0.0019 - val_acc: 0.7500 Epoch 1939/3000 1712/1712 [==============================] - 1s - loss: 8.5856e-04 - acc: 0.8183 - val_loss: 0.0017 - val_acc: 0.7360 Epoch 1940/3000 1712/1712 [==============================] - 1s - loss: 7.3568e-04 - acc: 0.8411 - val_loss: 0.0017 - val_acc: 0.7196 Epoch 1941/3000 1712/1712 [==============================] - 1s - loss: 8.3225e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7313 Epoch 1942/3000 1712/1712 [==============================] - 1s - loss: 7.7034e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 1943/3000 1712/1712 [==============================] - 1s - loss: 8.3042e-04 - acc: 0.8324 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 1944/3000 1712/1712 [==============================] - 1s - loss: 7.8837e-04 - acc: 0.8470 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 1945/3000 1712/1712 [==============================] - 1s - loss: 8.7791e-04 - acc: 0.8435 - val_loss: 0.0014 - val_acc: 0.7617 Epoch 1946/3000 1712/1712 [==============================] - 1s - loss: 6.6747e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7617 Epoch 1947/3000 1712/1712 [==============================] - 1s - loss: 9.0956e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1948/3000 1712/1712 [==============================] - 1s - loss: 7.4455e-04 - acc: 0.8452 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 1949/3000 1712/1712 [==============================] - 1s - loss: 7.8627e-04 - acc: 0.8300 - val_loss: 0.0019 - val_acc: 0.7453 Epoch 1950/3000 1712/1712 [==============================] - 1s - loss: 8.0959e-04 - acc: 0.8283 - val_loss: 0.0015 - val_acc: 0.7897 Epoch 1951/3000 1712/1712 [==============================] - 1s - loss: 8.2201e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7593 Epoch 1952/3000 1712/1712 [==============================] - 1s - loss: 8.1464e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 1953/3000 1712/1712 [==============================] - 1s - loss: 7.9905e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 1954/3000 1712/1712 [==============================] - 1s - loss: 8.3997e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 1955/3000 1712/1712 [==============================] - 1s - loss: 7.4123e-04 - acc: 0.8370 - val_loss: 0.0027 - val_acc: 0.7103 Epoch 1956/3000 1712/1712 [==============================] - 1s - loss: 8.2776e-04 - acc: 0.8324 - val_loss: 0.0017 - val_acc: 0.7570 Epoch 1957/3000 1712/1712 [==============================] - 1s - loss: 8.6112e-04 - acc: 0.8218 - val_loss: 0.0019 - val_acc: 0.7079 Epoch 1958/3000 1712/1712 [==============================] - 1s - loss: 7.3489e-04 - acc: 0.8382 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 1959/3000 1712/1712 [==============================] - 1s - loss: 8.3992e-04 - acc: 0.8429 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 1960/3000 1712/1712 [==============================] - 1s - loss: 8.0668e-04 - acc: 0.8370 - val_loss: 0.0014 - val_acc: 0.7640 Epoch 1961/3000 1712/1712 [==============================] - 1s - loss: 7.8065e-04 - acc: 0.8516 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1962/3000 1712/1712 [==============================] - 1s - loss: 9.1908e-04 - acc: 0.8189 - val_loss: 0.0014 - val_acc: 0.7407 Epoch 1963/3000 1712/1712 [==============================] - 1s - loss: 7.2269e-04 - acc: 0.8318 - val_loss: 0.0020 - val_acc: 0.7150 Epoch 1964/3000 1712/1712 [==============================] - 1s - loss: 8.3490e-04 - acc: 0.8248 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 1965/3000 1712/1712 [==============================] - 1s - loss: 7.9256e-04 - acc: 0.8435 - val_loss: 0.0024 - val_acc: 0.7266 Epoch 1966/3000 1712/1712 [==============================] - 1s - loss: 8.7283e-04 - acc: 0.8289 - val_loss: 0.0016 - val_acc: 0.7407 Epoch 1967/3000 1712/1712 [==============================] - 1s - loss: 7.4776e-04 - acc: 0.8470 - val_loss: 0.0017 - val_acc: 0.7290 Epoch 1968/3000 1712/1712 [==============================] - 1s - loss: 8.0872e-04 - acc: 0.8411 - val_loss: 0.0014 - val_acc: 0.7734 Epoch 1969/3000 1712/1712 [==============================] - 1s - loss: 8.2215e-04 - acc: 0.8458 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 1970/3000 1712/1712 [==============================] - 1s - loss: 8.0472e-04 - acc: 0.8154 - val_loss: 0.0012 - val_acc: 0.7570 Epoch 1971/3000 1712/1712 [==============================] - 1s - loss: 7.7232e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 1972/3000 1712/1712 [==============================] - 1s - loss: 8.1771e-04 - acc: 0.8312 - val_loss: 0.0017 - val_acc: 0.8014 Epoch 1973/3000 1712/1712 [==============================] - 1s - loss: 8.0958e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7850 Epoch 1974/3000 1712/1712 [==============================] - 1s - loss: 7.7931e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 1975/3000 1712/1712 [==============================] - 1s - loss: 7.7483e-04 - acc: 0.8505 - val_loss: 0.0014 - val_acc: 0.7874 Epoch 1976/3000 1712/1712 [==============================] - 1s - loss: 8.5199e-04 - acc: 0.8300 - val_loss: 0.0014 - val_acc: 0.7430 Epoch 1977/3000 1712/1712 [==============================] - 1s - loss: 7.6291e-04 - acc: 0.8464 - val_loss: 0.0013 - val_acc: 0.8131 Epoch 1978/3000 1712/1712 [==============================] - 1s - loss: 8.1654e-04 - acc: 0.8429 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 1979/3000 1712/1712 [==============================] - 1s - loss: 8.3227e-04 - acc: 0.8207 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 1980/3000 1712/1712 [==============================] - 1s - loss: 7.7139e-04 - acc: 0.8452 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 1981/3000 1712/1712 [==============================] - 1s - loss: 8.1170e-04 - acc: 0.8259 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 1982/3000 1712/1712 [==============================] - 1s - loss: 8.1544e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 1983/3000 1712/1712 [==============================] - 1s - loss: 7.8947e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 1984/3000 1712/1712 [==============================] - 1s - loss: 7.9002e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 1985/3000 1712/1712 [==============================] - 1s - loss: 8.1395e-04 - acc: 0.8359 - val_loss: 0.0014 - val_acc: 0.7570 Epoch 1986/3000 1712/1712 [==============================] - 1s - loss: 8.1949e-04 - acc: 0.8324 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 1987/3000 1712/1712 [==============================] - 1s - loss: 8.3605e-04 - acc: 0.8154 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 1988/3000 1712/1712 [==============================] - 1s - loss: 7.9293e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7593 Epoch 1989/3000 1712/1712 [==============================] - 1s - loss: 7.3670e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 1990/3000 1712/1712 [==============================] - 1s - loss: 7.7425e-04 - acc: 0.8271 - val_loss: 0.0022 - val_acc: 0.7360 Epoch 1991/3000 1712/1712 [==============================] - 1s - loss: 8.4203e-04 - acc: 0.8207 - val_loss: 0.0016 - val_acc: 0.7804 Epoch 1992/3000 1712/1712 [==============================] - 1s - loss: 8.1499e-04 - acc: 0.8370 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 1993/3000 1712/1712 [==============================] - 1s - loss: 8.2642e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 1994/3000 1712/1712 [==============================] - 1s - loss: 7.3632e-04 - acc: 0.8417 - val_loss: 0.0022 - val_acc: 0.7196 Epoch 1995/3000 1712/1712 [==============================] - 1s - loss: 8.2402e-04 - acc: 0.8324 - val_loss: 0.0019 - val_acc: 0.7523 Epoch 1996/3000 1712/1712 [==============================] - 1s - loss: 8.4117e-04 - acc: 0.8417 - val_loss: 0.0018 - val_acc: 0.7173 Epoch 1997/3000 1712/1712 [==============================] - 1s - loss: 8.0433e-04 - acc: 0.8382 - val_loss: 0.0019 - val_acc: 0.7617 Epoch 1998/3000 1712/1712 [==============================] - 1s - loss: 8.4481e-04 - acc: 0.8300 - val_loss: 0.0016 - val_acc: 0.7617 Epoch 1999/3000 1712/1712 [==============================] - 1s - loss: 7.9839e-04 - acc: 0.8400 - val_loss: 0.0016 - val_acc: 0.7991 Epoch 2000/3000 1712/1712 [==============================] - 1s - loss: 8.2029e-04 - acc: 0.8487 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 2001/3000 1712/1712 [==============================] - 1s - loss: 8.5753e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7593 Epoch 2002/3000 1712/1712 [==============================] - 1s - loss: 7.1847e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2003/3000 1712/1712 [==============================] - 1s - loss: 8.3374e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 2004/3000 1712/1712 [==============================] - 1s - loss: 8.4315e-04 - acc: 0.8382 - val_loss: 0.0020 - val_acc: 0.7477 Epoch 2005/3000 1712/1712 [==============================] - 1s - loss: 8.0735e-04 - acc: 0.8335 - val_loss: 0.0018 - val_acc: 0.7360 Epoch 2006/3000 1712/1712 [==============================] - 1s - loss: 8.1280e-04 - acc: 0.8306 - val_loss: 0.0019 - val_acc: 0.7336 Epoch 2007/3000 1712/1712 [==============================] - 1s - loss: 7.9089e-04 - acc: 0.8382 - val_loss: 0.0018 - val_acc: 0.7126 Epoch 2008/3000 1712/1712 [==============================] - 1s - loss: 7.7312e-04 - acc: 0.8429 - val_loss: 0.0022 - val_acc: 0.7430 Epoch 2009/3000 1712/1712 [==============================] - 1s - loss: 8.2265e-04 - acc: 0.8254 - val_loss: 0.0020 - val_acc: 0.7383 Epoch 2010/3000 1712/1712 [==============================] - 1s - loss: 8.3518e-04 - acc: 0.8242 - val_loss: 0.0019 - val_acc: 0.7687 Epoch 2011/3000 1712/1712 [==============================] - 1s - loss: 7.8701e-04 - acc: 0.8294 - val_loss: 0.0018 - val_acc: 0.7570 Epoch 2012/3000 1712/1712 [==============================] - 1s - loss: 8.0568e-04 - acc: 0.8265 - val_loss: 0.0025 - val_acc: 0.7126 Epoch 2013/3000 1712/1712 [==============================] - 1s - loss: 8.9646e-04 - acc: 0.8254 - val_loss: 0.0019 - val_acc: 0.7243 Epoch 2014/3000 1712/1712 [==============================] - 1s - loss: 7.7941e-04 - acc: 0.8324 - val_loss: 0.0018 - val_acc: 0.7570 Epoch 2015/3000 1712/1712 [==============================] - 1s - loss: 7.9979e-04 - acc: 0.8277 - val_loss: 0.0019 - val_acc: 0.7710 Epoch 2016/3000 1712/1712 [==============================] - 1s - loss: 8.5111e-04 - acc: 0.8271 - val_loss: 0.0019 - val_acc: 0.7290 Epoch 2017/3000 1712/1712 [==============================] - 1s - loss: 7.8999e-04 - acc: 0.8312 - val_loss: 0.0018 - val_acc: 0.7079 Epoch 2018/3000 1712/1712 [==============================] - 1s - loss: 7.9215e-04 - acc: 0.8254 - val_loss: 0.0015 - val_acc: 0.7523 Epoch 2019/3000 1712/1712 [==============================] - 1s - loss: 8.0379e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 2020/3000 1712/1712 [==============================] - 1s - loss: 8.2366e-04 - acc: 0.8230 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2021/3000 1712/1712 [==============================] - 1s - loss: 8.0919e-04 - acc: 0.8201 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 2022/3000 1712/1712 [==============================] - 1s - loss: 7.4925e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 2023/3000 1712/1712 [==============================] - 1s - loss: 7.8052e-04 - acc: 0.8417 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 2024/3000 1712/1712 [==============================] - 1s - loss: 8.4738e-04 - acc: 0.8341 - val_loss: 0.0018 - val_acc: 0.7290 Epoch 2025/3000 1712/1712 [==============================] - 1s - loss: 8.1088e-04 - acc: 0.8359 - val_loss: 0.0017 - val_acc: 0.7640 Epoch 2026/3000 1712/1712 [==============================] - 1s - loss: 8.0789e-04 - acc: 0.8364 - val_loss: 0.0015 - val_acc: 0.7944 Epoch 2027/3000 1712/1712 [==============================] - 1s - loss: 8.7279e-04 - acc: 0.8516 - val_loss: 0.0015 - val_acc: 0.7850 Epoch 2028/3000 1712/1712 [==============================] - 1s - loss: 7.9135e-04 - acc: 0.8324 - val_loss: 0.0017 - val_acc: 0.7477 Epoch 2029/3000 1712/1712 [==============================] - 1s - loss: 7.8530e-04 - acc: 0.8370 - val_loss: 0.0020 - val_acc: 0.7430 Epoch 2030/3000 1712/1712 [==============================] - 1s - loss: 8.1247e-04 - acc: 0.8353 - val_loss: 0.0019 - val_acc: 0.7383 Epoch 2031/3000 1712/1712 [==============================] - 1s - loss: 7.5595e-04 - acc: 0.8511 - val_loss: 0.0014 - val_acc: 0.7570 Epoch 2032/3000 1712/1712 [==============================] - 1s - loss: 8.4934e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 2033/3000 1712/1712 [==============================] - 1s - loss: 7.3368e-04 - acc: 0.8400 - val_loss: 0.0026 - val_acc: 0.7056 Epoch 2034/3000 1712/1712 [==============================] - 1s - loss: 9.1650e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2035/3000 1712/1712 [==============================] - 1s - loss: 7.7804e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 2036/3000 1712/1712 [==============================] - 1s - loss: 8.2831e-04 - acc: 0.8248 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 2037/3000 1712/1712 [==============================] - 1s - loss: 7.9187e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 2038/3000 1712/1712 [==============================] - 1s - loss: 8.1262e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 2039/3000 1712/1712 [==============================] - 1s - loss: 7.9588e-04 - acc: 0.8329 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 2040/3000 1712/1712 [==============================] - 1s - loss: 8.2083e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 2041/3000 1712/1712 [==============================] - 1s - loss: 8.2468e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2042/3000 1712/1712 [==============================] - 1s - loss: 8.3988e-04 - acc: 0.8435 - val_loss: 0.0014 - val_acc: 0.7850 Epoch 2043/3000 1712/1712 [==============================] - 1s - loss: 7.7179e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 2044/3000 1712/1712 [==============================] - 1s - loss: 7.9430e-04 - acc: 0.8353 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 2045/3000 1712/1712 [==============================] - 1s - loss: 7.9726e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.8201 Epoch 2046/3000 1712/1712 [==============================] - 1s - loss: 8.5062e-04 - acc: 0.8125 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 2047/3000 1712/1712 [==============================] - 1s - loss: 7.7812e-04 - acc: 0.8452 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 2048/3000 1712/1712 [==============================] - 1s - loss: 7.9945e-04 - acc: 0.8236 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2049/3000 1712/1712 [==============================] - 1s - loss: 8.0232e-04 - acc: 0.8265 - val_loss: 0.0022 - val_acc: 0.7103 Epoch 2050/3000 1712/1712 [==============================] - 1s - loss: 8.2608e-04 - acc: 0.8411 - val_loss: 0.0016 - val_acc: 0.7383 Epoch 2051/3000 1712/1712 [==============================] - 1s - loss: 8.1144e-04 - acc: 0.8248 - val_loss: 0.0027 - val_acc: 0.7313 Epoch 2052/3000 1712/1712 [==============================] - 1s - loss: 8.0422e-04 - acc: 0.8300 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 2053/3000 1712/1712 [==============================] - 1s - loss: 8.6349e-04 - acc: 0.8452 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 2054/3000 1712/1712 [==============================] - 1s - loss: 7.8191e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 2055/3000 1712/1712 [==============================] - 1s - loss: 8.1900e-04 - acc: 0.8475 - val_loss: 0.0015 - val_acc: 0.7453 Epoch 2056/3000 1712/1712 [==============================] - 1s - loss: 7.2622e-04 - acc: 0.8435 - val_loss: 0.0019 - val_acc: 0.7290 Epoch 2057/3000 1712/1712 [==============================] - 1s - loss: 8.4239e-04 - acc: 0.8446 - val_loss: 0.0016 - val_acc: 0.7804 Epoch 2058/3000 1712/1712 [==============================] - 1s - loss: 8.6492e-04 - acc: 0.8586 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 2059/3000 1712/1712 [==============================] - 1s - loss: 7.8807e-04 - acc: 0.8289 - val_loss: 0.0017 - val_acc: 0.7313 Epoch 2060/3000 1712/1712 [==============================] - 1s - loss: 8.9350e-04 - acc: 0.8137 - val_loss: 0.0019 - val_acc: 0.7593 Epoch 2061/3000 1712/1712 [==============================] - 1s - loss: 7.2504e-04 - acc: 0.8312 - val_loss: 0.0022 - val_acc: 0.7173 Epoch 2062/3000 1712/1712 [==============================] - 1s - loss: 8.3028e-04 - acc: 0.8400 - val_loss: 0.0019 - val_acc: 0.7266 Epoch 2063/3000 1712/1712 [==============================] - 1s - loss: 8.8872e-04 - acc: 0.8283 - val_loss: 0.0018 - val_acc: 0.7500 Epoch 2064/3000 1712/1712 [==============================] - 1s - loss: 7.5243e-04 - acc: 0.8353 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 2065/3000 1712/1712 [==============================] - 1s - loss: 8.2843e-04 - acc: 0.8300 - val_loss: 0.0020 - val_acc: 0.7407 Epoch 2066/3000 1712/1712 [==============================] - 1s - loss: 7.8847e-04 - acc: 0.8411 - val_loss: 0.0017 - val_acc: 0.7407 Epoch 2067/3000 1712/1712 [==============================] - 1s - loss: 8.0823e-04 - acc: 0.8359 - val_loss: 0.0015 - val_acc: 0.8037 Epoch 2068/3000 1712/1712 [==============================] - 1s - loss: 7.7715e-04 - acc: 0.8400 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2069/3000 1712/1712 [==============================] - 1s - loss: 6.8322e-04 - acc: 0.8452 - val_loss: 0.0021 - val_acc: 0.7290 Epoch 2070/3000 1712/1712 [==============================] - 1s - loss: 9.7441e-04 - acc: 0.8224 - val_loss: 0.0022 - val_acc: 0.7523 Epoch 2071/3000 1712/1712 [==============================] - 1s - loss: 7.6077e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 2072/3000 1712/1712 [==============================] - 1s - loss: 8.0978e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 2073/3000 1712/1712 [==============================] - 1s - loss: 8.1480e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 2074/3000 1712/1712 [==============================] - 1s - loss: 7.9333e-04 - acc: 0.8481 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2075/3000 1712/1712 [==============================] - 1s - loss: 8.7588e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 2076/3000 1712/1712 [==============================] - 1s - loss: 7.3286e-04 - acc: 0.8335 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2077/3000 1712/1712 [==============================] - 1s - loss: 8.5566e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 2078/3000 1712/1712 [==============================] - 1s - loss: 7.6532e-04 - acc: 0.8452 - val_loss: 0.0011 - val_acc: 0.8014 Epoch 2079/3000 1712/1712 [==============================] - 1s - loss: 8.6091e-04 - acc: 0.8224 - val_loss: 0.0014 - val_acc: 0.7570 Epoch 2080/3000 1712/1712 [==============================] - 1s - loss: 8.3163e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.8061 Epoch 2081/3000 1712/1712 [==============================] - 1s - loss: 7.4913e-04 - acc: 0.8283 - val_loss: 0.0015 - val_acc: 0.7570 Epoch 2082/3000 1712/1712 [==============================] - 1s - loss: 7.7702e-04 - acc: 0.8277 - val_loss: 0.0025 - val_acc: 0.7243 Epoch 2083/3000 1712/1712 [==============================] - 1s - loss: 8.3163e-04 - acc: 0.8283 - val_loss: 0.0018 - val_acc: 0.7407 Epoch 2084/3000 1712/1712 [==============================] - 1s - loss: 8.4584e-04 - acc: 0.8160 - val_loss: 0.0016 - val_acc: 0.7664 Epoch 2085/3000 1712/1712 [==============================] - 1s - loss: 7.3483e-04 - acc: 0.8429 - val_loss: 0.0022 - val_acc: 0.7173 Epoch 2086/3000 1712/1712 [==============================] - 1s - loss: 7.8005e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2087/3000 1712/1712 [==============================] - 1s - loss: 8.4871e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 2088/3000 1712/1712 [==============================] - 1s - loss: 7.8465e-04 - acc: 0.8364 - val_loss: 0.0011 - val_acc: 0.8107 Epoch 2089/3000 1712/1712 [==============================] - 1s - loss: 8.2146e-04 - acc: 0.8400 - val_loss: 0.0015 - val_acc: 0.7710 Epoch 2090/3000 1712/1712 [==============================] - 1s - loss: 8.0713e-04 - acc: 0.8172 - val_loss: 0.0020 - val_acc: 0.7640 Epoch 2091/3000 1712/1712 [==============================] - 1s - loss: 8.1804e-04 - acc: 0.8400 - val_loss: 0.0022 - val_acc: 0.7407 Epoch 2092/3000 1712/1712 [==============================] - 1s - loss: 7.9350e-04 - acc: 0.8440 - val_loss: 0.0024 - val_acc: 0.7407 Epoch 2093/3000 1712/1712 [==============================] - 1s - loss: 8.0258e-04 - acc: 0.8376 - val_loss: 0.0021 - val_acc: 0.7453 Epoch 2094/3000 1712/1712 [==============================] - 1s - loss: 8.4143e-04 - acc: 0.8230 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 2095/3000 1712/1712 [==============================] - 1s - loss: 7.9629e-04 - acc: 0.8440 - val_loss: 0.0022 - val_acc: 0.7780 Epoch 2096/3000 1712/1712 [==============================] - 1s - loss: 7.9675e-04 - acc: 0.8277 - val_loss: 0.0017 - val_acc: 0.7593 Epoch 2097/3000 1712/1712 [==============================] - 1s - loss: 8.1517e-04 - acc: 0.8353 - val_loss: 0.0020 - val_acc: 0.7150 Epoch 2098/3000 1712/1712 [==============================] - 1s - loss: 8.2833e-04 - acc: 0.8364 - val_loss: 0.0020 - val_acc: 0.7336 Epoch 2099/3000 1712/1712 [==============================] - 1s - loss: 7.3750e-04 - acc: 0.8400 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2100/3000 1712/1712 [==============================] - 1s - loss: 9.1112e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 2101/3000 1712/1712 [==============================] - 1s - loss: 7.7924e-04 - acc: 0.8493 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2102/3000 1712/1712 [==============================] - 1s - loss: 8.2176e-04 - acc: 0.8563 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 2103/3000 1712/1712 [==============================] - 1s - loss: 7.9594e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 2104/3000 1712/1712 [==============================] - 1s - loss: 9.3439e-04 - acc: 0.8172 - val_loss: 0.0014 - val_acc: 0.7617 Epoch 2105/3000 1712/1712 [==============================] - 1s - loss: 7.5728e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 2106/3000 1712/1712 [==============================] - 1s - loss: 7.6363e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 2107/3000 1712/1712 [==============================] - 1s - loss: 7.9027e-04 - acc: 0.8475 - val_loss: 0.0016 - val_acc: 0.7780 Epoch 2108/3000 1712/1712 [==============================] - 1s - loss: 8.1989e-04 - acc: 0.8364 - val_loss: 0.0017 - val_acc: 0.7570 Epoch 2109/3000 1712/1712 [==============================] - 1s - loss: 8.0788e-04 - acc: 0.8300 - val_loss: 0.0020 - val_acc: 0.7196 Epoch 2110/3000 1712/1712 [==============================] - 1s - loss: 8.5696e-04 - acc: 0.8259 - val_loss: 0.0017 - val_acc: 0.7383 Epoch 2111/3000 1712/1712 [==============================] - 1s - loss: 7.7678e-04 - acc: 0.8359 - val_loss: 0.0018 - val_acc: 0.7383 Epoch 2112/3000 1712/1712 [==============================] - 1s - loss: 7.8088e-04 - acc: 0.8207 - val_loss: 0.0021 - val_acc: 0.7290 Epoch 2113/3000 1712/1712 [==============================] - 1s - loss: 8.2733e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 2114/3000 1712/1712 [==============================] - 1s - loss: 7.7814e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 2115/3000 1712/1712 [==============================] - 1s - loss: 8.3021e-04 - acc: 0.8400 - val_loss: 0.0014 - val_acc: 0.7477 Epoch 2116/3000 1712/1712 [==============================] - 1s - loss: 8.6332e-04 - acc: 0.8341 - val_loss: 0.0018 - val_acc: 0.7196 Epoch 2117/3000 1712/1712 [==============================] - 1s - loss: 7.6807e-04 - acc: 0.8347 - val_loss: 0.0020 - val_acc: 0.7336 Epoch 2118/3000 1712/1712 [==============================] - 1s - loss: 7.9239e-04 - acc: 0.8341 - val_loss: 0.0020 - val_acc: 0.7407 Epoch 2119/3000 1712/1712 [==============================] - 1s - loss: 8.1253e-04 - acc: 0.8329 - val_loss: 0.0016 - val_acc: 0.7664 Epoch 2120/3000 1712/1712 [==============================] - 1s - loss: 8.0749e-04 - acc: 0.8335 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 2121/3000 1712/1712 [==============================] - 1s - loss: 8.1483e-04 - acc: 0.8218 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 2122/3000 1712/1712 [==============================] - 1s - loss: 7.7745e-04 - acc: 0.8487 - val_loss: 0.0011 - val_acc: 0.8084 Epoch 2123/3000 1712/1712 [==============================] - 1s - loss: 8.3253e-04 - acc: 0.8289 - val_loss: 0.0014 - val_acc: 0.7827 Epoch 2124/3000 1712/1712 [==============================] - 1s - loss: 7.7836e-04 - acc: 0.8417 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 2125/3000 1712/1712 [==============================] - 1s - loss: 8.3980e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 2126/3000 1712/1712 [==============================] - 1s - loss: 8.0638e-04 - acc: 0.8347 - val_loss: 0.0017 - val_acc: 0.7757 Epoch 2127/3000 1712/1712 [==============================] - 1s - loss: 6.8195e-04 - acc: 0.8405 - val_loss: 0.0022 - val_acc: 0.7126 Epoch 2128/3000 1712/1712 [==============================] - 1s - loss: 9.0105e-04 - acc: 0.8359 - val_loss: 0.0015 - val_acc: 0.7734 Epoch 2129/3000 1712/1712 [==============================] - 1s - loss: 8.8855e-04 - acc: 0.8452 - val_loss: 0.0017 - val_acc: 0.7500 Epoch 2130/3000 1712/1712 [==============================] - 1s - loss: 7.4045e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2131/3000 1712/1712 [==============================] - 1s - loss: 8.3297e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 2132/3000 1712/1712 [==============================] - 1s - loss: 7.3511e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2133/3000 1712/1712 [==============================] - 1s - loss: 8.2486e-04 - acc: 0.8435 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 2134/3000 1712/1712 [==============================] - 1s - loss: 8.4960e-04 - acc: 0.8259 - val_loss: 0.0017 - val_acc: 0.7243 Epoch 2135/3000 1712/1712 [==============================] - 1s - loss: 7.4987e-04 - acc: 0.8201 - val_loss: 0.0020 - val_acc: 0.7430 Epoch 2136/3000 1712/1712 [==============================] - 1s - loss: 8.3114e-04 - acc: 0.8388 - val_loss: 0.0021 - val_acc: 0.7150 Epoch 2137/3000 1712/1712 [==============================] - 1s - loss: 8.4810e-04 - acc: 0.8213 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 2138/3000 1712/1712 [==============================] - 1s - loss: 7.6027e-04 - acc: 0.8248 - val_loss: 0.0018 - val_acc: 0.7640 Epoch 2139/3000 1712/1712 [==============================] - 1s - loss: 8.2686e-04 - acc: 0.8271 - val_loss: 0.0020 - val_acc: 0.7453 Epoch 2140/3000 1712/1712 [==============================] - 1s - loss: 8.5761e-04 - acc: 0.8446 - val_loss: 0.0017 - val_acc: 0.7383 Epoch 2141/3000 1712/1712 [==============================] - 1s - loss: 6.9796e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2142/3000 1712/1712 [==============================] - 1s - loss: 8.8415e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 2143/3000 1712/1712 [==============================] - 1s - loss: 7.9421e-04 - acc: 0.8242 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 2144/3000 1712/1712 [==============================] - 1s - loss: 8.3271e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 2145/3000 1712/1712 [==============================] - 1s - loss: 8.3707e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 2146/3000 1712/1712 [==============================] - 1s - loss: 6.9679e-04 - acc: 0.8440 - val_loss: 0.0011 - val_acc: 0.7944 Epoch 2147/3000 1712/1712 [==============================] - 1s - loss: 9.2095e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 2148/3000 1712/1712 [==============================] - 1s - loss: 7.5060e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.7664 Epoch 2149/3000 1712/1712 [==============================] - 1s - loss: 8.2304e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 2150/3000 1712/1712 [==============================] - 1s - loss: 8.1209e-04 - acc: 0.8405 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 2151/3000 1712/1712 [==============================] - 1s - loss: 7.7599e-04 - acc: 0.8370 - val_loss: 0.0013 - val_acc: 0.7593 Epoch 2152/3000 1712/1712 [==============================] - 1s - loss: 7.9208e-04 - acc: 0.8440 - val_loss: 0.0020 - val_acc: 0.7500 Epoch 2153/3000 1712/1712 [==============================] - 1s - loss: 7.9313e-04 - acc: 0.8271 - val_loss: 0.0019 - val_acc: 0.7220 Epoch 2154/3000 1712/1712 [==============================] - 1s - loss: 8.5648e-04 - acc: 0.8382 - val_loss: 0.0016 - val_acc: 0.7523 Epoch 2155/3000 1712/1712 [==============================] - 1s - loss: 8.0039e-04 - acc: 0.8353 - val_loss: 0.0014 - val_acc: 0.7734 Epoch 2156/3000 1712/1712 [==============================] - 1s - loss: 8.0703e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2157/3000 1712/1712 [==============================] - 1s - loss: 7.7415e-04 - acc: 0.8324 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 2158/3000 1712/1712 [==============================] - 1s - loss: 8.4431e-04 - acc: 0.8335 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 2159/3000 1712/1712 [==============================] - 1s - loss: 8.3136e-04 - acc: 0.8218 - val_loss: 0.0021 - val_acc: 0.7243 Epoch 2160/3000 1712/1712 [==============================] - 1s - loss: 7.3015e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 2161/3000 1712/1712 [==============================] - 1s - loss: 8.7930e-04 - acc: 0.8242 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 2162/3000 1712/1712 [==============================] - 1s - loss: 8.5453e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2163/3000 1712/1712 [==============================] - 1s - loss: 7.3638e-04 - acc: 0.8353 - val_loss: 0.0014 - val_acc: 0.7757 Epoch 2164/3000 1712/1712 [==============================] - 1s - loss: 8.0233e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 2165/3000 1712/1712 [==============================] - 1s - loss: 8.3897e-04 - acc: 0.8195 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 2166/3000 1712/1712 [==============================] - 1s - loss: 8.0930e-04 - acc: 0.8435 - val_loss: 0.0013 - val_acc: 0.7640 Epoch 2167/3000 1712/1712 [==============================] - 1s - loss: 7.8644e-04 - acc: 0.8300 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 2168/3000 1712/1712 [==============================] - 1s - loss: 7.6711e-04 - acc: 0.8172 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 2169/3000 1712/1712 [==============================] - 1s - loss: 8.0381e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 2170/3000 1712/1712 [==============================] - 1s - loss: 8.1332e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 2171/3000 1712/1712 [==============================] - 1s - loss: 7.3807e-04 - acc: 0.8481 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 2172/3000 1712/1712 [==============================] - 1s - loss: 8.5460e-04 - acc: 0.8148 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 2173/3000 1712/1712 [==============================] - 1s - loss: 7.6191e-04 - acc: 0.8505 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2174/3000 1712/1712 [==============================] - 1s - loss: 7.9532e-04 - acc: 0.8353 - val_loss: 0.0014 - val_acc: 0.8131 Epoch 2175/3000 1712/1712 [==============================] - 1s - loss: 8.4455e-04 - acc: 0.8248 - val_loss: 0.0014 - val_acc: 0.7664 Epoch 2176/3000 1712/1712 [==============================] - 1s - loss: 7.2849e-04 - acc: 0.8324 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 2177/3000 1712/1712 [==============================] - 1s - loss: 8.3804e-04 - acc: 0.8435 - val_loss: 0.0014 - val_acc: 0.7944 Epoch 2178/3000 1712/1712 [==============================] - 1s - loss: 8.1325e-04 - acc: 0.8341 - val_loss: 0.0020 - val_acc: 0.7523 Epoch 2179/3000 1712/1712 [==============================] - 1s - loss: 8.0055e-04 - acc: 0.8394 - val_loss: 0.0020 - val_acc: 0.7336 Epoch 2180/3000 1712/1712 [==============================] - 1s - loss: 8.4463e-04 - acc: 0.8341 - val_loss: 0.0017 - val_acc: 0.7664 Epoch 2181/3000 1712/1712 [==============================] - 1s - loss: 7.5739e-04 - acc: 0.8429 - val_loss: 0.0013 - val_acc: 0.8131 Epoch 2182/3000 1712/1712 [==============================] - 1s - loss: 8.7626e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 2183/3000 1712/1712 [==============================] - 1s - loss: 8.1150e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2184/3000 1712/1712 [==============================] - 1s - loss: 8.0625e-04 - acc: 0.8254 - val_loss: 0.0014 - val_acc: 0.7967 Epoch 2185/3000 1712/1712 [==============================] - 1s - loss: 7.5509e-04 - acc: 0.8417 - val_loss: 0.0018 - val_acc: 0.7617 Epoch 2186/3000 1712/1712 [==============================] - 1s - loss: 9.4426e-04 - acc: 0.8178 - val_loss: 0.0018 - val_acc: 0.7664 Epoch 2187/3000 1712/1712 [==============================] - 1s - loss: 7.5815e-04 - acc: 0.8382 - val_loss: 0.0015 - val_acc: 0.7477 Epoch 2188/3000 1712/1712 [==============================] - 1s - loss: 8.1490e-04 - acc: 0.8353 - val_loss: 0.0011 - val_acc: 0.7780 Epoch 2189/3000 1712/1712 [==============================] - 1s - loss: 8.5757e-04 - acc: 0.8435 - val_loss: 0.0015 - val_acc: 0.7383 Epoch 2190/3000 1712/1712 [==============================] - 1s - loss: 8.2364e-04 - acc: 0.8353 - val_loss: 0.0018 - val_acc: 0.7523 Epoch 2191/3000 1712/1712 [==============================] - 1s - loss: 7.8471e-04 - acc: 0.8405 - val_loss: 0.0018 - val_acc: 0.7430 Epoch 2192/3000 1712/1712 [==============================] - 1s - loss: 7.9254e-04 - acc: 0.8318 - val_loss: 0.0014 - val_acc: 0.7617 Epoch 2193/3000 1712/1712 [==============================] - 1s - loss: 8.2312e-04 - acc: 0.8230 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 2194/3000 1712/1712 [==============================] - 1s - loss: 8.7459e-04 - acc: 0.8364 - val_loss: 0.0013 - val_acc: 0.7500 Epoch 2195/3000 1712/1712 [==============================] - 1s - loss: 7.1979e-04 - acc: 0.8505 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2196/3000 1712/1712 [==============================] - 1s - loss: 8.4106e-04 - acc: 0.8364 - val_loss: 0.0013 - val_acc: 0.8061 Epoch 2197/3000 1712/1712 [==============================] - 1s - loss: 8.2022e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 2198/3000 1712/1712 [==============================] - 1s - loss: 8.1355e-04 - acc: 0.8394 - val_loss: 0.0016 - val_acc: 0.7523 Epoch 2199/3000 1712/1712 [==============================] - 1s - loss: 8.2886e-04 - acc: 0.8341 - val_loss: 0.0020 - val_acc: 0.7477 Epoch 2200/3000 1712/1712 [==============================] - 1s - loss: 8.1850e-04 - acc: 0.8283 - val_loss: 0.0018 - val_acc: 0.7827 Epoch 2201/3000 1712/1712 [==============================] - 1s - loss: 8.1277e-04 - acc: 0.8277 - val_loss: 0.0018 - val_acc: 0.7967 Epoch 2202/3000 1712/1712 [==============================] - 1s - loss: 7.8039e-04 - acc: 0.8242 - val_loss: 0.0017 - val_acc: 0.7336 Epoch 2203/3000 1712/1712 [==============================] - 1s - loss: 7.9742e-04 - acc: 0.8429 - val_loss: 0.0021 - val_acc: 0.7220 Epoch 2204/3000 1712/1712 [==============================] - 1s - loss: 7.8450e-04 - acc: 0.8359 - val_loss: 0.0015 - val_acc: 0.7336 Epoch 2205/3000 1712/1712 [==============================] - 1s - loss: 8.8470e-04 - acc: 0.8236 - val_loss: 0.0016 - val_acc: 0.7407 Epoch 2206/3000 1712/1712 [==============================] - 1s - loss: 7.4372e-04 - acc: 0.8423 - val_loss: 0.0024 - val_acc: 0.7103 Epoch 2207/3000 1712/1712 [==============================] - 1s - loss: 7.9612e-04 - acc: 0.8300 - val_loss: 0.0015 - val_acc: 0.7710 Epoch 2208/3000 1712/1712 [==============================] - 1s - loss: 8.1841e-04 - acc: 0.8376 - val_loss: 0.0018 - val_acc: 0.7523 Epoch 2209/3000 1712/1712 [==============================] - 1s - loss: 8.0915e-04 - acc: 0.8113 - val_loss: 0.0016 - val_acc: 0.7407 Epoch 2210/3000 1712/1712 [==============================] - 1s - loss: 8.6090e-04 - acc: 0.8364 - val_loss: 0.0017 - val_acc: 0.7430 Epoch 2211/3000 1712/1712 [==============================] - 1s - loss: 8.5071e-04 - acc: 0.8370 - val_loss: 0.0021 - val_acc: 0.7360 Epoch 2212/3000 1712/1712 [==============================] - 1s - loss: 7.5844e-04 - acc: 0.8458 - val_loss: 0.0017 - val_acc: 0.7266 Epoch 2213/3000 1712/1712 [==============================] - 1s - loss: 7.9493e-04 - acc: 0.8289 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 2214/3000 1712/1712 [==============================] - 1s - loss: 7.4541e-04 - acc: 0.8435 - val_loss: 0.0012 - val_acc: 0.7640 Epoch 2215/3000 1712/1712 [==============================] - 1s - loss: 8.0939e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 2216/3000 1712/1712 [==============================] - 1s - loss: 8.0997e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 2217/3000 1712/1712 [==============================] - 1s - loss: 8.6524e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 2218/3000 1712/1712 [==============================] - 1s - loss: 7.5801e-04 - acc: 0.8440 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2219/3000 1712/1712 [==============================] - 1s - loss: 8.0322e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.7593 Epoch 2220/3000 1712/1712 [==============================] - 1s - loss: 8.1343e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 2221/3000 1712/1712 [==============================] - 1s - loss: 7.3069e-04 - acc: 0.8329 - val_loss: 0.0014 - val_acc: 0.7757 Epoch 2222/3000 1712/1712 [==============================] - 1s - loss: 7.9913e-04 - acc: 0.8242 - val_loss: 0.0012 - val_acc: 0.7640 Epoch 2223/3000 1712/1712 [==============================] - 1s - loss: 8.0885e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2224/3000 1712/1712 [==============================] - 1s - loss: 8.0648e-04 - acc: 0.8435 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 2225/3000 1712/1712 [==============================] - 1s - loss: 8.1605e-04 - acc: 0.8364 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2226/3000 1712/1712 [==============================] - 1s - loss: 7.5936e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 2227/3000 1712/1712 [==============================] - 1s - loss: 8.4569e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2228/3000 1712/1712 [==============================] - 1s - loss: 7.7030e-04 - acc: 0.8353 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 2229/3000 1712/1712 [==============================] - 1s - loss: 8.0975e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 2230/3000 1712/1712 [==============================] - 1s - loss: 8.6219e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2231/3000 1712/1712 [==============================] - 1s - loss: 7.6003e-04 - acc: 0.8405 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 2232/3000 1712/1712 [==============================] - 1s - loss: 8.2445e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 2233/3000 1712/1712 [==============================] - 1s - loss: 7.7305e-04 - acc: 0.8505 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 2234/3000 1712/1712 [==============================] - 1s - loss: 8.0909e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7523 Epoch 2235/3000 1712/1712 [==============================] - 1s - loss: 7.8079e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2236/3000 1712/1712 [==============================] - 1s - loss: 7.8117e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 2237/3000 1712/1712 [==============================] - 1s - loss: 7.3230e-04 - acc: 0.8388 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 2238/3000 1712/1712 [==============================] - 1s - loss: 9.3677e-04 - acc: 0.8201 - val_loss: 0.0013 - val_acc: 0.7336 Epoch 2239/3000 1712/1712 [==============================] - 1s - loss: 7.0732e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 2240/3000 1712/1712 [==============================] - 1s - loss: 8.1480e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 2241/3000 1712/1712 [==============================] - 1s - loss: 7.7116e-04 - acc: 0.8435 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2242/3000 1712/1712 [==============================] - 1s - loss: 8.7965e-04 - acc: 0.8189 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2243/3000 1712/1712 [==============================] - 1s - loss: 7.6430e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 2244/3000 1712/1712 [==============================] - 1s - loss: 7.9227e-04 - acc: 0.8283 - val_loss: 0.0011 - val_acc: 0.8131 Epoch 2245/3000 1712/1712 [==============================] - 1s - loss: 7.3749e-04 - acc: 0.8440 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2246/3000 1712/1712 [==============================] - 1s - loss: 8.9066e-04 - acc: 0.8289 - val_loss: 0.0021 - val_acc: 0.7570 Epoch 2247/3000 1712/1712 [==============================] - 1s - loss: 7.9954e-04 - acc: 0.8394 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 2248/3000 1712/1712 [==============================] - 1s - loss: 7.8212e-04 - acc: 0.8446 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 2249/3000 1712/1712 [==============================] - 1s - loss: 9.0372e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 2250/3000 1712/1712 [==============================] - 1s - loss: 6.6709e-04 - acc: 0.8370 - val_loss: 0.0018 - val_acc: 0.7874 Epoch 2251/3000 1712/1712 [==============================] - 1s - loss: 8.6071e-04 - acc: 0.8329 - val_loss: 0.0019 - val_acc: 0.7243 Epoch 2252/3000 1712/1712 [==============================] - 1s - loss: 7.3993e-04 - acc: 0.8370 - val_loss: 0.0019 - val_acc: 0.7547 Epoch 2253/3000 1712/1712 [==============================] - 1s - loss: 7.9938e-04 - acc: 0.8493 - val_loss: 0.0018 - val_acc: 0.7313 Epoch 2254/3000 1712/1712 [==============================] - 1s - loss: 8.1787e-04 - acc: 0.8364 - val_loss: 0.0019 - val_acc: 0.7336 Epoch 2255/3000 1712/1712 [==============================] - 1s - loss: 9.5003e-04 - acc: 0.8201 - val_loss: 0.0017 - val_acc: 0.7196 Epoch 2256/3000 1712/1712 [==============================] - 1s - loss: 7.9290e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 2257/3000 1712/1712 [==============================] - 1s - loss: 7.8582e-04 - acc: 0.8277 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 2258/3000 1712/1712 [==============================] - 1s - loss: 8.1429e-04 - acc: 0.8411 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 2259/3000 1712/1712 [==============================] - 1s - loss: 8.2171e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 2260/3000 1712/1712 [==============================] - 1s - loss: 7.8894e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.8131 Epoch 2261/3000 1712/1712 [==============================] - 1s - loss: 8.3602e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2262/3000 1712/1712 [==============================] - 1s - loss: 7.7592e-04 - acc: 0.8312 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 2263/3000 1712/1712 [==============================] - 1s - loss: 8.2505e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 2264/3000 1712/1712 [==============================] - 1s - loss: 7.5310e-04 - acc: 0.8335 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 2265/3000 1712/1712 [==============================] - 1s - loss: 7.6933e-04 - acc: 0.8440 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 2266/3000 1712/1712 [==============================] - 1s - loss: 7.8199e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 2267/3000 1712/1712 [==============================] - 1s - loss: 8.4953e-04 - acc: 0.8417 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 2268/3000 1712/1712 [==============================] - 1s - loss: 8.1154e-04 - acc: 0.8324 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 2269/3000 1712/1712 [==============================] - 1s - loss: 7.7409e-04 - acc: 0.8446 - val_loss: 0.0017 - val_acc: 0.7687 Epoch 2270/3000 1712/1712 [==============================] - 1s - loss: 8.4126e-04 - acc: 0.8195 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 2271/3000 1712/1712 [==============================] - 1s - loss: 7.4204e-04 - acc: 0.8499 - val_loss: 0.0017 - val_acc: 0.7336 Epoch 2272/3000 1712/1712 [==============================] - 1s - loss: 8.0820e-04 - acc: 0.8259 - val_loss: 0.0018 - val_acc: 0.7897 Epoch 2273/3000 1712/1712 [==============================] - 1s - loss: 8.1534e-04 - acc: 0.8516 - val_loss: 0.0020 - val_acc: 0.7547 Epoch 2274/3000 1712/1712 [==============================] - 1s - loss: 8.1753e-04 - acc: 0.8411 - val_loss: 0.0014 - val_acc: 0.7850 Epoch 2275/3000 1712/1712 [==============================] - 1s - loss: 7.7473e-04 - acc: 0.8394 - val_loss: 0.0015 - val_acc: 0.7897 Epoch 2276/3000 1712/1712 [==============================] - 1s - loss: 8.7010e-04 - acc: 0.8230 - val_loss: 0.0017 - val_acc: 0.7757 Epoch 2277/3000 1712/1712 [==============================] - 1s - loss: 8.2500e-04 - acc: 0.8294 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 2278/3000 1712/1712 [==============================] - 1s - loss: 7.9961e-04 - acc: 0.8359 - val_loss: 0.0016 - val_acc: 0.7617 Epoch 2279/3000 1712/1712 [==============================] - 1s - loss: 8.0494e-04 - acc: 0.8277 - val_loss: 0.0017 - val_acc: 0.7897 Epoch 2280/3000 1712/1712 [==============================] - 1s - loss: 7.7971e-04 - acc: 0.8563 - val_loss: 0.0018 - val_acc: 0.7640 Epoch 2281/3000 1712/1712 [==============================] - 1s - loss: 8.7981e-04 - acc: 0.8277 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 2282/3000 1712/1712 [==============================] - 1s - loss: 7.6536e-04 - acc: 0.8318 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 2283/3000 1712/1712 [==============================] - 1s - loss: 7.9084e-04 - acc: 0.8300 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 2284/3000 1712/1712 [==============================] - 1s - loss: 7.8531e-04 - acc: 0.8400 - val_loss: 0.0016 - val_acc: 0.8014 Epoch 2285/3000 1712/1712 [==============================] - 1s - loss: 8.1768e-04 - acc: 0.8382 - val_loss: 0.0021 - val_acc: 0.7360 Epoch 2286/3000 1712/1712 [==============================] - 1s - loss: 7.8562e-04 - acc: 0.8405 - val_loss: 0.0019 - val_acc: 0.7547 Epoch 2287/3000 1712/1712 [==============================] - 1s - loss: 7.9557e-04 - acc: 0.8283 - val_loss: 0.0016 - val_acc: 0.7383 Epoch 2288/3000 1712/1712 [==============================] - 1s - loss: 8.5182e-04 - acc: 0.8353 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 2289/3000 1712/1712 [==============================] - 1s - loss: 7.8628e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2290/3000 1712/1712 [==============================] - 1s - loss: 7.8796e-04 - acc: 0.8423 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 2291/3000 1712/1712 [==============================] - 1s - loss: 8.1437e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 2292/3000 1712/1712 [==============================] - 1s - loss: 8.3176e-04 - acc: 0.8341 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 2293/3000 1712/1712 [==============================] - 1s - loss: 8.0487e-04 - acc: 0.8218 - val_loss: 0.0014 - val_acc: 0.7617 Epoch 2294/3000 1712/1712 [==============================] - 1s - loss: 7.6178e-04 - acc: 0.8207 - val_loss: 0.0018 - val_acc: 0.7757 Epoch 2295/3000 1712/1712 [==============================] - 1s - loss: 8.4795e-04 - acc: 0.8294 - val_loss: 0.0017 - val_acc: 0.7430 Epoch 2296/3000 1712/1712 [==============================] - 1s - loss: 7.9607e-04 - acc: 0.8440 - val_loss: 0.0018 - val_acc: 0.7547 Epoch 2297/3000 1712/1712 [==============================] - 1s - loss: 8.0537e-04 - acc: 0.8242 - val_loss: 0.0017 - val_acc: 0.7944 Epoch 2298/3000 1712/1712 [==============================] - 1s - loss: 7.9387e-04 - acc: 0.8324 - val_loss: 0.0021 - val_acc: 0.7407 Epoch 2299/3000 1712/1712 [==============================] - 1s - loss: 8.3399e-04 - acc: 0.8277 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 2300/3000 1712/1712 [==============================] - 1s - loss: 7.7577e-04 - acc: 0.8248 - val_loss: 0.0017 - val_acc: 0.7196 Epoch 2301/3000 1712/1712 [==============================] - 1s - loss: 7.5174e-04 - acc: 0.8411 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2302/3000 1712/1712 [==============================] - 1s - loss: 8.5041e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.8271 Epoch 2303/3000 1712/1712 [==============================] - 1s - loss: 8.1226e-04 - acc: 0.8440 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 2304/3000 1712/1712 [==============================] - 1s - loss: 7.7756e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2305/3000 1712/1712 [==============================] - 1s - loss: 8.6268e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 2306/3000 1712/1712 [==============================] - 1s - loss: 7.9521e-04 - acc: 0.8324 - val_loss: 0.0014 - val_acc: 0.7266 Epoch 2307/3000 1712/1712 [==============================] - 1s - loss: 7.9444e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.8224 Epoch 2308/3000 1712/1712 [==============================] - 1s - loss: 7.9242e-04 - acc: 0.8370 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 2309/3000 1712/1712 [==============================] - 1s - loss: 7.5592e-04 - acc: 0.8370 - val_loss: 0.0014 - val_acc: 0.7827 Epoch 2310/3000 1712/1712 [==============================] - 1s - loss: 8.4458e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 2311/3000 1712/1712 [==============================] - 1s - loss: 7.5734e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 2312/3000 1712/1712 [==============================] - 1s - loss: 8.9109e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.8061 Epoch 2313/3000 1712/1712 [==============================] - 1s - loss: 7.6829e-04 - acc: 0.8376 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 2314/3000 1712/1712 [==============================] - 1s - loss: 8.4215e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 2315/3000 1712/1712 [==============================] - 1s - loss: 7.4853e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 2316/3000 1712/1712 [==============================] - 1s - loss: 7.3826e-04 - acc: 0.8411 - val_loss: 0.0015 - val_acc: 0.7453 Epoch 2317/3000 1712/1712 [==============================] - 1s - loss: 6.7495e-04 - acc: 0.8493 - val_loss: 0.0011 - val_acc: 0.7967 Epoch 2318/3000 1712/1712 [==============================] - 1s - loss: 9.5613e-04 - acc: 0.8458 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2319/3000 1712/1712 [==============================] - 1s - loss: 8.6008e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 2320/3000 1712/1712 [==============================] - 1s - loss: 6.9246e-04 - acc: 0.8475 - val_loss: 0.0014 - val_acc: 0.7523 Epoch 2321/3000 1712/1712 [==============================] - 1s - loss: 9.2057e-04 - acc: 0.8277 - val_loss: 0.0015 - val_acc: 0.7407 Epoch 2322/3000 1712/1712 [==============================] - 1s - loss: 7.5794e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7967 Epoch 2323/3000 1712/1712 [==============================] - 1s - loss: 8.0410e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 2324/3000 1712/1712 [==============================] - 1s - loss: 8.0845e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 2325/3000 1712/1712 [==============================] - 1s - loss: 7.9831e-04 - acc: 0.8376 - val_loss: 0.0012 - val_acc: 0.8201 Epoch 2326/3000 1712/1712 [==============================] - 1s - loss: 8.2977e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 2327/3000 1712/1712 [==============================] - 1s - loss: 7.4804e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2328/3000 1712/1712 [==============================] - 1s - loss: 8.5261e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 2329/3000 1712/1712 [==============================] - 1s - loss: 7.1848e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 2330/3000 1712/1712 [==============================] - 1s - loss: 8.1435e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 2331/3000 1712/1712 [==============================] - 1s - loss: 8.2858e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 2332/3000 1712/1712 [==============================] - 1s - loss: 7.8792e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 2333/3000 1712/1712 [==============================] - 1s - loss: 8.9570e-04 - acc: 0.8458 - val_loss: 0.0014 - val_acc: 0.7874 Epoch 2334/3000 1712/1712 [==============================] - 1s - loss: 6.6518e-04 - acc: 0.8493 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 2335/3000 1712/1712 [==============================] - 1s - loss: 8.0044e-04 - acc: 0.8306 - val_loss: 0.0022 - val_acc: 0.7220 Epoch 2336/3000 1712/1712 [==============================] - 1s - loss: 9.4240e-04 - acc: 0.8195 - val_loss: 0.0015 - val_acc: 0.7921 Epoch 2337/3000 1712/1712 [==============================] - 1s - loss: 7.6686e-04 - acc: 0.8435 - val_loss: 0.0017 - val_acc: 0.7453 Epoch 2338/3000 1712/1712 [==============================] - 1s - loss: 8.3410e-04 - acc: 0.8271 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 2339/3000 1712/1712 [==============================] - 1s - loss: 8.0623e-04 - acc: 0.8312 - val_loss: 0.0017 - val_acc: 0.7617 Epoch 2340/3000 1712/1712 [==============================] - 1s - loss: 7.9077e-04 - acc: 0.8388 - val_loss: 0.0011 - val_acc: 0.8131 Epoch 2341/3000 1712/1712 [==============================] - 1s - loss: 8.2452e-04 - acc: 0.8376 - val_loss: 0.0015 - val_acc: 0.7874 Epoch 2342/3000 1712/1712 [==============================] - 1s - loss: 7.0223e-04 - acc: 0.8464 - val_loss: 0.0018 - val_acc: 0.7593 Epoch 2343/3000 1712/1712 [==============================] - 1s - loss: 8.2511e-04 - acc: 0.8277 - val_loss: 0.0018 - val_acc: 0.7500 Epoch 2344/3000 1712/1712 [==============================] - 1s - loss: 8.2016e-04 - acc: 0.8370 - val_loss: 0.0018 - val_acc: 0.7383 Epoch 2345/3000 1712/1712 [==============================] - 1s - loss: 7.6674e-04 - acc: 0.8294 - val_loss: 0.0022 - val_acc: 0.7243 Epoch 2346/3000 1712/1712 [==============================] - 1s - loss: 8.2845e-04 - acc: 0.8376 - val_loss: 0.0023 - val_acc: 0.7056 Epoch 2347/3000 1712/1712 [==============================] - 1s - loss: 7.3916e-04 - acc: 0.8388 - val_loss: 0.0014 - val_acc: 0.7593 Epoch 2348/3000 1712/1712 [==============================] - 1s - loss: 9.3581e-04 - acc: 0.8294 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 2349/3000 1712/1712 [==============================] - 1s - loss: 7.4029e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 2350/3000 1712/1712 [==============================] - 1s - loss: 8.0946e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 2351/3000 1712/1712 [==============================] - 1s - loss: 8.1786e-04 - acc: 0.8300 - val_loss: 0.0016 - val_acc: 0.7710 Epoch 2352/3000 1712/1712 [==============================] - 1s - loss: 7.7132e-04 - acc: 0.8440 - val_loss: 0.0023 - val_acc: 0.7243 Epoch 2353/3000 1712/1712 [==============================] - 1s - loss: 8.3846e-04 - acc: 0.8329 - val_loss: 0.0017 - val_acc: 0.7547 Epoch 2354/3000 1712/1712 [==============================] - 1s - loss: 8.0877e-04 - acc: 0.8429 - val_loss: 0.0016 - val_acc: 0.7477 Epoch 2355/3000 1712/1712 [==============================] - 1s - loss: 8.6917e-04 - acc: 0.8324 - val_loss: 0.0011 - val_acc: 0.8107 Epoch 2356/3000 1712/1712 [==============================] - 1s - loss: 7.1878e-04 - acc: 0.8435 - val_loss: 0.0011 - val_acc: 0.8037 Epoch 2357/3000 1712/1712 [==============================] - 1s - loss: 8.4790e-04 - acc: 0.8376 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 2358/3000 1712/1712 [==============================] - 1s - loss: 7.5402e-04 - acc: 0.8481 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 2359/3000 1712/1712 [==============================] - 1s - loss: 8.5943e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2360/3000 1712/1712 [==============================] - 1s - loss: 8.2390e-04 - acc: 0.8324 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2361/3000 1712/1712 [==============================] - 1s - loss: 7.8408e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 2362/3000 1712/1712 [==============================] - 1s - loss: 7.6491e-04 - acc: 0.8347 - val_loss: 0.0014 - val_acc: 0.8037 Epoch 2363/3000 1712/1712 [==============================] - 1s - loss: 8.1738e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2364/3000 1712/1712 [==============================] - 1s - loss: 7.4480e-04 - acc: 0.8242 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 2365/3000 1712/1712 [==============================] - 1s - loss: 8.1914e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 2366/3000 1712/1712 [==============================] - 1s - loss: 7.8659e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.8201 Epoch 2367/3000 1712/1712 [==============================] - 1s - loss: 7.9434e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 2368/3000 1712/1712 [==============================] - 1s - loss: 7.6862e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 2369/3000 1712/1712 [==============================] - 1s - loss: 8.0980e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 2370/3000 1712/1712 [==============================] - 1s - loss: 7.6297e-04 - acc: 0.8411 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 2371/3000 1712/1712 [==============================] - 1s - loss: 8.4114e-04 - acc: 0.8224 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 2372/3000 1712/1712 [==============================] - 1s - loss: 7.4891e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2373/3000 1712/1712 [==============================] - 1s - loss: 7.9981e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 2374/3000 1712/1712 [==============================] - 1s - loss: 8.7543e-04 - acc: 0.8429 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 2375/3000 1712/1712 [==============================] - 1s - loss: 7.3017e-04 - acc: 0.8347 - val_loss: 0.0014 - val_acc: 0.7991 Epoch 2376/3000 1712/1712 [==============================] - 1s - loss: 8.1275e-04 - acc: 0.8417 - val_loss: 0.0019 - val_acc: 0.7173 Epoch 2377/3000 1712/1712 [==============================] - 1s - loss: 7.9564e-04 - acc: 0.8370 - val_loss: 0.0018 - val_acc: 0.7430 Epoch 2378/3000 1712/1712 [==============================] - 1s - loss: 7.1974e-04 - acc: 0.8318 - val_loss: 0.0025 - val_acc: 0.7056 Epoch 2379/3000 1712/1712 [==============================] - 1s - loss: 8.1334e-04 - acc: 0.8376 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 2380/3000 1712/1712 [==============================] - 1s - loss: 8.8606e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 2381/3000 1712/1712 [==============================] - 1s - loss: 7.7504e-04 - acc: 0.8353 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 2382/3000 1712/1712 [==============================] - 1s - loss: 8.1691e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 2383/3000 1712/1712 [==============================] - 1s - loss: 8.3233e-04 - acc: 0.8271 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 2384/3000 1712/1712 [==============================] - 1s - loss: 8.0402e-04 - acc: 0.8300 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 2385/3000 1712/1712 [==============================] - 1s - loss: 7.5395e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.7640 Epoch 2386/3000 1712/1712 [==============================] - 1s - loss: 7.9227e-04 - acc: 0.8464 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 2387/3000 1712/1712 [==============================] - 1s - loss: 7.6351e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 2388/3000 1712/1712 [==============================] - 1s - loss: 7.4933e-04 - acc: 0.8364 - val_loss: 0.0011 - val_acc: 0.8061 Epoch 2389/3000 1712/1712 [==============================] - 1s - loss: 8.3798e-04 - acc: 0.8254 - val_loss: 0.0014 - val_acc: 0.7523 Epoch 2390/3000 1712/1712 [==============================] - 1s - loss: 8.1115e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2391/3000 1712/1712 [==============================] - 1s - loss: 7.7652e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 2392/3000 1712/1712 [==============================] - 1s - loss: 8.1406e-04 - acc: 0.8417 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2393/3000 1712/1712 [==============================] - 1s - loss: 8.0234e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 2394/3000 1712/1712 [==============================] - 1s - loss: 7.4807e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 2395/3000 1712/1712 [==============================] - 1s - loss: 7.9839e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 2396/3000 1712/1712 [==============================] - 1s - loss: 8.0958e-04 - acc: 0.8405 - val_loss: 0.0019 - val_acc: 0.7383 Epoch 2397/3000 1712/1712 [==============================] - 1s - loss: 8.3861e-04 - acc: 0.8195 - val_loss: 0.0020 - val_acc: 0.7243 Epoch 2398/3000 1712/1712 [==============================] - 1s - loss: 7.7380e-04 - acc: 0.8388 - val_loss: 0.0018 - val_acc: 0.7734 Epoch 2399/3000 1712/1712 [==============================] - 1s - loss: 8.9830e-04 - acc: 0.8370 - val_loss: 0.0018 - val_acc: 0.7734 Epoch 2400/3000 1712/1712 [==============================] - 1s - loss: 7.8676e-04 - acc: 0.8458 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 2401/3000 1712/1712 [==============================] - 1s - loss: 7.9830e-04 - acc: 0.8277 - val_loss: 0.0017 - val_acc: 0.7360 Epoch 2402/3000 1712/1712 [==============================] - 1s - loss: 8.4530e-04 - acc: 0.8411 - val_loss: 0.0017 - val_acc: 0.7664 Epoch 2403/3000 1712/1712 [==============================] - 1s - loss: 7.5316e-04 - acc: 0.8283 - val_loss: 0.0019 - val_acc: 0.7593 Epoch 2404/3000 1712/1712 [==============================] - 1s - loss: 8.2390e-04 - acc: 0.8300 - val_loss: 0.0015 - val_acc: 0.7407 Epoch 2405/3000 1712/1712 [==============================] - 1s - loss: 7.8328e-04 - acc: 0.8201 - val_loss: 0.0019 - val_acc: 0.7874 Epoch 2406/3000 1712/1712 [==============================] - 1s - loss: 7.7684e-04 - acc: 0.8283 - val_loss: 0.0021 - val_acc: 0.7196 Epoch 2407/3000 1712/1712 [==============================] - 1s - loss: 7.9160e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2408/3000 1712/1712 [==============================] - 1s - loss: 8.0507e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 2409/3000 1712/1712 [==============================] - 1s - loss: 8.2030e-04 - acc: 0.8335 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 2410/3000 1712/1712 [==============================] - 1s - loss: 7.4035e-04 - acc: 0.8440 - val_loss: 0.0013 - val_acc: 0.7944 Epoch 2411/3000 1712/1712 [==============================] - 1s - loss: 7.9835e-04 - acc: 0.8400 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 2412/3000 1712/1712 [==============================] - 1s - loss: 8.2384e-04 - acc: 0.8259 - val_loss: 0.0016 - val_acc: 0.7500 Epoch 2413/3000 1712/1712 [==============================] - 1s - loss: 8.0295e-04 - acc: 0.8405 - val_loss: 0.0019 - val_acc: 0.7266 Epoch 2414/3000 1712/1712 [==============================] - 1s - loss: 7.5276e-04 - acc: 0.8324 - val_loss: 0.0021 - val_acc: 0.7196 Epoch 2415/3000 1712/1712 [==============================] - 1s - loss: 8.6285e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2416/3000 1712/1712 [==============================] - 1s - loss: 8.4102e-04 - acc: 0.8335 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 2417/3000 1712/1712 [==============================] - 1s - loss: 8.1615e-04 - acc: 0.8178 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 2418/3000 1712/1712 [==============================] - 1s - loss: 7.4985e-04 - acc: 0.8411 - val_loss: 0.0015 - val_acc: 0.7477 Epoch 2419/3000 1712/1712 [==============================] - 1s - loss: 8.8877e-04 - acc: 0.8364 - val_loss: 0.0021 - val_acc: 0.7430 Epoch 2420/3000 1712/1712 [==============================] - 1s - loss: 7.9651e-04 - acc: 0.8312 - val_loss: 0.0020 - val_acc: 0.7617 Epoch 2421/3000 1712/1712 [==============================] - 1s - loss: 7.7625e-04 - acc: 0.8329 - val_loss: 0.0017 - val_acc: 0.8061 Epoch 2422/3000 1712/1712 [==============================] - 1s - loss: 7.6422e-04 - acc: 0.8359 - val_loss: 0.0019 - val_acc: 0.7710 Epoch 2423/3000 1712/1712 [==============================] - 1s - loss: 8.5063e-04 - acc: 0.8329 - val_loss: 0.0020 - val_acc: 0.7593 Epoch 2424/3000 1712/1712 [==============================] - 1s - loss: 8.4711e-04 - acc: 0.8230 - val_loss: 0.0021 - val_acc: 0.7243 Epoch 2425/3000 1712/1712 [==============================] - 1s - loss: 7.4853e-04 - acc: 0.8364 - val_loss: 0.0018 - val_acc: 0.7593 Epoch 2426/3000 1712/1712 [==============================] - 1s - loss: 8.8149e-04 - acc: 0.8271 - val_loss: 0.0016 - val_acc: 0.7687 Epoch 2427/3000 1712/1712 [==============================] - 1s - loss: 8.3023e-04 - acc: 0.8481 - val_loss: 0.0015 - val_acc: 0.7827 Epoch 2428/3000 1712/1712 [==============================] - 1s - loss: 7.6397e-04 - acc: 0.8475 - val_loss: 0.0018 - val_acc: 0.7290 Epoch 2429/3000 1712/1712 [==============================] - 1s - loss: 7.4659e-04 - acc: 0.8388 - val_loss: 0.0021 - val_acc: 0.7196 Epoch 2430/3000 1712/1712 [==============================] - 1s - loss: 8.1320e-04 - acc: 0.8265 - val_loss: 0.0020 - val_acc: 0.7313 Epoch 2431/3000 1712/1712 [==============================] - 1s - loss: 8.2134e-04 - acc: 0.8522 - val_loss: 0.0022 - val_acc: 0.7453 Epoch 2432/3000 1712/1712 [==============================] - 1s - loss: 8.6367e-04 - acc: 0.8271 - val_loss: 0.0015 - val_acc: 0.7547 Epoch 2433/3000 1712/1712 [==============================] - 1s - loss: 8.7275e-04 - acc: 0.8277 - val_loss: 0.0015 - val_acc: 0.7664 Epoch 2434/3000 1712/1712 [==============================] - 1s - loss: 7.2765e-04 - acc: 0.8271 - val_loss: 0.0025 - val_acc: 0.7079 Epoch 2435/3000 1712/1712 [==============================] - 1s - loss: 8.5100e-04 - acc: 0.8347 - val_loss: 0.0016 - val_acc: 0.7827 Epoch 2436/3000 1712/1712 [==============================] - 1s - loss: 8.1141e-04 - acc: 0.8347 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 2437/3000 1712/1712 [==============================] - 1s - loss: 8.3258e-04 - acc: 0.8394 - val_loss: 0.0013 - val_acc: 0.8061 Epoch 2438/3000 1712/1712 [==============================] - 1s - loss: 7.5851e-04 - acc: 0.8370 - val_loss: 0.0016 - val_acc: 0.7407 Epoch 2439/3000 1712/1712 [==============================] - 1s - loss: 8.4626e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 2440/3000 1712/1712 [==============================] - 1s - loss: 7.9744e-04 - acc: 0.8335 - val_loss: 0.0018 - val_acc: 0.7827 Epoch 2441/3000 1712/1712 [==============================] - 1s - loss: 8.3991e-04 - acc: 0.8271 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 2442/3000 1712/1712 [==============================] - 1s - loss: 7.9143e-04 - acc: 0.8347 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 2443/3000 1712/1712 [==============================] - 1s - loss: 7.5361e-04 - acc: 0.8312 - val_loss: 0.0015 - val_acc: 0.8014 Epoch 2444/3000 1712/1712 [==============================] - 1s - loss: 8.0069e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 2445/3000 1712/1712 [==============================] - 1s - loss: 7.8286e-04 - acc: 0.8400 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 2446/3000 1712/1712 [==============================] - 1s - loss: 8.5508e-04 - acc: 0.8405 - val_loss: 0.0016 - val_acc: 0.7336 Epoch 2447/3000 1712/1712 [==============================] - 1s - loss: 7.3802e-04 - acc: 0.8511 - val_loss: 0.0021 - val_acc: 0.7500 Epoch 2448/3000 1712/1712 [==============================] - 1s - loss: 8.3081e-04 - acc: 0.8341 - val_loss: 0.0021 - val_acc: 0.7079 Epoch 2449/3000 1712/1712 [==============================] - 1s - loss: 8.0007e-04 - acc: 0.8329 - val_loss: 0.0019 - val_acc: 0.7710 Epoch 2450/3000 1712/1712 [==============================] - 1s - loss: 8.0744e-04 - acc: 0.8470 - val_loss: 0.0016 - val_acc: 0.7407 Epoch 2451/3000 1712/1712 [==============================] - 1s - loss: 8.1688e-04 - acc: 0.8458 - val_loss: 0.0014 - val_acc: 0.7734 Epoch 2452/3000 1712/1712 [==============================] - 1s - loss: 7.6330e-04 - acc: 0.8481 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 2453/3000 1712/1712 [==============================] - 1s - loss: 7.9852e-04 - acc: 0.8318 - val_loss: 0.0016 - val_acc: 0.7383 Epoch 2454/3000 1712/1712 [==============================] - 1s - loss: 8.4646e-04 - acc: 0.8189 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 2455/3000 1712/1712 [==============================] - 1s - loss: 8.0439e-04 - acc: 0.8364 - val_loss: 0.0017 - val_acc: 0.7617 Epoch 2456/3000 1712/1712 [==============================] - 1s - loss: 7.8657e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.8154 Epoch 2457/3000 1712/1712 [==============================] - 1s - loss: 8.2105e-04 - acc: 0.8230 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2458/3000 1712/1712 [==============================] - 1s - loss: 7.4819e-04 - acc: 0.8318 - val_loss: 0.0015 - val_acc: 0.7640 Epoch 2459/3000 1712/1712 [==============================] - 1s - loss: 8.9805e-04 - acc: 0.8405 - val_loss: 0.0012 - val_acc: 0.8201 Epoch 2460/3000 1712/1712 [==============================] - 1s - loss: 7.1282e-04 - acc: 0.8271 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2461/3000 1712/1712 [==============================] - 1s - loss: 8.0930e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 2462/3000 1712/1712 [==============================] - 1s - loss: 7.9747e-04 - acc: 0.8400 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 2463/3000 1712/1712 [==============================] - 1s - loss: 7.7094e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 2464/3000 1712/1712 [==============================] - 1s - loss: 7.7646e-04 - acc: 0.8405 - val_loss: 0.0020 - val_acc: 0.7593 Epoch 2465/3000 1712/1712 [==============================] - 1s - loss: 8.1914e-04 - acc: 0.8359 - val_loss: 0.0016 - val_acc: 0.7547 Epoch 2466/3000 1712/1712 [==============================] - 1s - loss: 8.4957e-04 - acc: 0.8306 - val_loss: 0.0015 - val_acc: 0.7617 Epoch 2467/3000 1712/1712 [==============================] - 1s - loss: 7.4517e-04 - acc: 0.8563 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 2468/3000 1712/1712 [==============================] - 1s - loss: 7.9709e-04 - acc: 0.8382 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 2469/3000 1712/1712 [==============================] - 1s - loss: 8.5514e-04 - acc: 0.8248 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2470/3000 1712/1712 [==============================] - 1s - loss: 7.9935e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7874 Epoch 2471/3000 1712/1712 [==============================] - 1s - loss: 7.8004e-04 - acc: 0.8394 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 2472/3000 1712/1712 [==============================] - 1s - loss: 7.7029e-04 - acc: 0.8511 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 2473/3000 1712/1712 [==============================] - 1s - loss: 8.8974e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 2474/3000 1712/1712 [==============================] - 1s - loss: 8.3261e-04 - acc: 0.8440 - val_loss: 0.0012 - val_acc: 0.7640 Epoch 2475/3000 1712/1712 [==============================] - 1s - loss: 7.9819e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2476/3000 1712/1712 [==============================] - 1s - loss: 7.8301e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 2477/3000 1712/1712 [==============================] - 1s - loss: 7.4062e-04 - acc: 0.8364 - val_loss: 0.0014 - val_acc: 0.8061 Epoch 2478/3000 1712/1712 [==============================] - 1s - loss: 8.3264e-04 - acc: 0.8224 - val_loss: 0.0015 - val_acc: 0.7710 Epoch 2479/3000 1712/1712 [==============================] - 1s - loss: 7.5782e-04 - acc: 0.8277 - val_loss: 0.0012 - val_acc: 0.7687 Epoch 2480/3000 1712/1712 [==============================] - 1s - loss: 8.0114e-04 - acc: 0.8224 - val_loss: 0.0014 - val_acc: 0.7874 Epoch 2481/3000 1712/1712 [==============================] - 1s - loss: 8.3549e-04 - acc: 0.8411 - val_loss: 0.0025 - val_acc: 0.7453 Epoch 2482/3000 1712/1712 [==============================] - 1s - loss: 7.8733e-04 - acc: 0.8329 - val_loss: 0.0021 - val_acc: 0.7523 Epoch 2483/3000 1712/1712 [==============================] - 1s - loss: 8.3832e-04 - acc: 0.8446 - val_loss: 0.0017 - val_acc: 0.7477 Epoch 2484/3000 1712/1712 [==============================] - 1s - loss: 8.1615e-04 - acc: 0.8318 - val_loss: 0.0020 - val_acc: 0.7383 Epoch 2485/3000 1712/1712 [==============================] - 1s - loss: 7.8358e-04 - acc: 0.8359 - val_loss: 0.0017 - val_acc: 0.7547 Epoch 2486/3000 1712/1712 [==============================] - 1s - loss: 8.1673e-04 - acc: 0.8411 - val_loss: 0.0017 - val_acc: 0.7874 Epoch 2487/3000 1712/1712 [==============================] - 1s - loss: 7.7237e-04 - acc: 0.8259 - val_loss: 0.0014 - val_acc: 0.7430 Epoch 2488/3000 1712/1712 [==============================] - 1s - loss: 8.3195e-04 - acc: 0.8265 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 2489/3000 1712/1712 [==============================] - 1s - loss: 7.7946e-04 - acc: 0.8329 - val_loss: 0.0019 - val_acc: 0.7804 Epoch 2490/3000 1712/1712 [==============================] - 1s - loss: 8.1275e-04 - acc: 0.8446 - val_loss: 0.0020 - val_acc: 0.7383 Epoch 2491/3000 1712/1712 [==============================] - 1s - loss: 7.9734e-04 - acc: 0.8435 - val_loss: 0.0019 - val_acc: 0.7336 Epoch 2492/3000 1712/1712 [==============================] - 1s - loss: 8.3245e-04 - acc: 0.8364 - val_loss: 0.0021 - val_acc: 0.7336 Epoch 2493/3000 1712/1712 [==============================] - 1s - loss: 7.8056e-04 - acc: 0.8335 - val_loss: 0.0011 - val_acc: 0.8178 Epoch 2494/3000 1712/1712 [==============================] - 1s - loss: 7.8748e-04 - acc: 0.8464 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 2495/3000 1712/1712 [==============================] - 1s - loss: 7.9195e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2496/3000 1712/1712 [==============================] - 1s - loss: 8.1636e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 2497/3000 1712/1712 [==============================] - 1s - loss: 8.0140e-04 - acc: 0.8446 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 2498/3000 1712/1712 [==============================] - 1s - loss: 8.3414e-04 - acc: 0.8324 - val_loss: 0.0020 - val_acc: 0.7196 Epoch 2499/3000 1712/1712 [==============================] - 1s - loss: 7.4911e-04 - acc: 0.8382 - val_loss: 0.0017 - val_acc: 0.7593 Epoch 2500/3000 1712/1712 [==============================] - 1s - loss: 8.1560e-04 - acc: 0.8464 - val_loss: 0.0022 - val_acc: 0.7430 Epoch 2501/3000 1712/1712 [==============================] - 1s - loss: 8.0326e-04 - acc: 0.8335 - val_loss: 0.0015 - val_acc: 0.7593 Epoch 2502/3000 1712/1712 [==============================] - 1s - loss: 8.0573e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 2503/3000 1712/1712 [==============================] - 1s - loss: 7.8219e-04 - acc: 0.8277 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 2504/3000 1712/1712 [==============================] - 1s - loss: 8.4207e-04 - acc: 0.8417 - val_loss: 0.0020 - val_acc: 0.7056 Epoch 2505/3000 1712/1712 [==============================] - 1s - loss: 7.6673e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 2506/3000 1712/1712 [==============================] - 1s - loss: 7.8519e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2507/3000 1712/1712 [==============================] - 1s - loss: 8.0347e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 2508/3000 1712/1712 [==============================] - 1s - loss: 8.5426e-04 - acc: 0.8218 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2509/3000 1712/1712 [==============================] - 1s - loss: 7.7234e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 2510/3000 1712/1712 [==============================] - 1s - loss: 7.0814e-04 - acc: 0.8499 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 2511/3000 1712/1712 [==============================] - 1s - loss: 8.2625e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2512/3000 1712/1712 [==============================] - 1s - loss: 8.7682e-04 - acc: 0.8429 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 2513/3000 1712/1712 [==============================] - 1s - loss: 7.2611e-04 - acc: 0.8423 - val_loss: 0.0011 - val_acc: 0.7967 Epoch 2514/3000 1712/1712 [==============================] - 1s - loss: 7.9353e-04 - acc: 0.8376 - val_loss: 0.0014 - val_acc: 0.7617 Epoch 2515/3000 1712/1712 [==============================] - 1s - loss: 8.1435e-04 - acc: 0.8289 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 2516/3000 1712/1712 [==============================] - 1s - loss: 8.2750e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 2517/3000 1712/1712 [==============================] - 1s - loss: 7.7282e-04 - acc: 0.8353 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 2518/3000 1712/1712 [==============================] - 1s - loss: 7.9367e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 2519/3000 1712/1712 [==============================] - 1s - loss: 7.8729e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 2520/3000 1712/1712 [==============================] - 1s - loss: 7.5816e-04 - acc: 0.8364 - val_loss: 0.0015 - val_acc: 0.7967 Epoch 2521/3000 1712/1712 [==============================] - 1s - loss: 8.6568e-04 - acc: 0.8458 - val_loss: 0.0021 - val_acc: 0.7477 Epoch 2522/3000 1712/1712 [==============================] - 1s - loss: 7.1181e-04 - acc: 0.8464 - val_loss: 0.0019 - val_acc: 0.7430 Epoch 2523/3000 1712/1712 [==============================] - 1s - loss: 8.5869e-04 - acc: 0.8242 - val_loss: 0.0017 - val_acc: 0.7196 Epoch 2524/3000 1712/1712 [==============================] - 1s - loss: 8.0695e-04 - acc: 0.8376 - val_loss: 0.0016 - val_acc: 0.7640 Epoch 2525/3000 1712/1712 [==============================] - 1s - loss: 8.2765e-04 - acc: 0.8370 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2526/3000 1712/1712 [==============================] - 1s - loss: 6.8115e-04 - acc: 0.8458 - val_loss: 0.0014 - val_acc: 0.7523 Epoch 2527/3000 1712/1712 [==============================] - 1s - loss: 8.5205e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7547 Epoch 2528/3000 1712/1712 [==============================] - 1s - loss: 7.8774e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2529/3000 1712/1712 [==============================] - 1s - loss: 6.8186e-04 - acc: 0.8452 - val_loss: 0.0011 - val_acc: 0.7734 Epoch 2530/3000 1712/1712 [==============================] - 1s - loss: 9.0932e-04 - acc: 0.8254 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 2531/3000 1712/1712 [==============================] - 1s - loss: 7.8443e-04 - acc: 0.8218 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2532/3000 1712/1712 [==============================] - 1s - loss: 8.1688e-04 - acc: 0.8370 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 2533/3000 1712/1712 [==============================] - 1s - loss: 8.1687e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7734 Epoch 2534/3000 1712/1712 [==============================] - 1s - loss: 7.6096e-04 - acc: 0.8522 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 2535/3000 1712/1712 [==============================] - 1s - loss: 7.9991e-04 - acc: 0.8189 - val_loss: 0.0017 - val_acc: 0.7430 Epoch 2536/3000 1712/1712 [==============================] - 1s - loss: 8.2726e-04 - acc: 0.8364 - val_loss: 0.0021 - val_acc: 0.7477 Epoch 2537/3000 1712/1712 [==============================] - 1s - loss: 8.0039e-04 - acc: 0.8283 - val_loss: 0.0017 - val_acc: 0.7897 Epoch 2538/3000 1712/1712 [==============================] - 1s - loss: 8.1773e-04 - acc: 0.8359 - val_loss: 0.0017 - val_acc: 0.7617 Epoch 2539/3000 1712/1712 [==============================] - 1s - loss: 8.2318e-04 - acc: 0.8277 - val_loss: 0.0018 - val_acc: 0.7290 Epoch 2540/3000 1712/1712 [==============================] - 1s - loss: 7.9177e-04 - acc: 0.8242 - val_loss: 0.0015 - val_acc: 0.7453 Epoch 2541/3000 1712/1712 [==============================] - 1s - loss: 8.2161e-04 - acc: 0.8335 - val_loss: 0.0021 - val_acc: 0.7780 Epoch 2542/3000 1712/1712 [==============================] - 1s - loss: 8.1260e-04 - acc: 0.8329 - val_loss: 0.0020 - val_acc: 0.7336 Epoch 2543/3000 1712/1712 [==============================] - 1s - loss: 7.8235e-04 - acc: 0.8335 - val_loss: 0.0020 - val_acc: 0.7360 Epoch 2544/3000 1712/1712 [==============================] - 1s - loss: 8.7781e-04 - acc: 0.8236 - val_loss: 0.0020 - val_acc: 0.7079 Epoch 2545/3000 1712/1712 [==============================] - 1s - loss: 8.0376e-04 - acc: 0.8405 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 2546/3000 1712/1712 [==============================] - 1s - loss: 7.8251e-04 - acc: 0.8172 - val_loss: 0.0023 - val_acc: 0.7243 Epoch 2547/3000 1712/1712 [==============================] - 1s - loss: 8.2857e-04 - acc: 0.8312 - val_loss: 0.0021 - val_acc: 0.7664 Epoch 2548/3000 1712/1712 [==============================] - 1s - loss: 7.7802e-04 - acc: 0.8364 - val_loss: 0.0017 - val_acc: 0.7874 Epoch 2549/3000 1712/1712 [==============================] - 1s - loss: 8.3555e-04 - acc: 0.8265 - val_loss: 0.0016 - val_acc: 0.7617 Epoch 2550/3000 1712/1712 [==============================] - 1s - loss: 8.3931e-04 - acc: 0.8213 - val_loss: 0.0023 - val_acc: 0.7243 Epoch 2551/3000 1712/1712 [==============================] - 1s - loss: 7.2481e-04 - acc: 0.8411 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 2552/3000 1712/1712 [==============================] - 1s - loss: 9.2444e-04 - acc: 0.8283 - val_loss: 0.0018 - val_acc: 0.7430 Epoch 2553/3000 1712/1712 [==============================] - 1s - loss: 7.5027e-04 - acc: 0.8429 - val_loss: 0.0018 - val_acc: 0.7850 Epoch 2554/3000 1712/1712 [==============================] - 1s - loss: 8.1673e-04 - acc: 0.8289 - val_loss: 0.0018 - val_acc: 0.7523 Epoch 2555/3000 1712/1712 [==============================] - 1s - loss: 8.0572e-04 - acc: 0.8335 - val_loss: 0.0020 - val_acc: 0.7734 Epoch 2556/3000 1712/1712 [==============================] - 1s - loss: 7.8706e-04 - acc: 0.8341 - val_loss: 0.0018 - val_acc: 0.7313 Epoch 2557/3000 1712/1712 [==============================] - 1s - loss: 8.1757e-04 - acc: 0.8277 - val_loss: 0.0024 - val_acc: 0.7196 Epoch 2558/3000 1712/1712 [==============================] - 1s - loss: 8.6674e-04 - acc: 0.8201 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 2559/3000 1712/1712 [==============================] - 1s - loss: 7.7718e-04 - acc: 0.8458 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 2560/3000 1712/1712 [==============================] - 1s - loss: 8.0974e-04 - acc: 0.8353 - val_loss: 0.0015 - val_acc: 0.7617 Epoch 2561/3000 1712/1712 [==============================] - 1s - loss: 7.8817e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 2562/3000 1712/1712 [==============================] - 1s - loss: 7.8514e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 2563/3000 1712/1712 [==============================] - 1s - loss: 7.6714e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 2564/3000 1712/1712 [==============================] - 1s - loss: 8.2773e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 2565/3000 1712/1712 [==============================] - 1s - loss: 7.4371e-04 - acc: 0.8224 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2566/3000 1712/1712 [==============================] - 1s - loss: 8.0235e-04 - acc: 0.8329 - val_loss: 0.0021 - val_acc: 0.7430 Epoch 2567/3000 1712/1712 [==============================] - 1s - loss: 7.3508e-04 - acc: 0.8411 - val_loss: 0.0012 - val_acc: 0.8154 Epoch 2568/3000 1712/1712 [==============================] - 1s - loss: 8.4830e-04 - acc: 0.8423 - val_loss: 0.0013 - val_acc: 0.7640 Epoch 2569/3000 1712/1712 [==============================] - 1s - loss: 7.1023e-04 - acc: 0.8475 - val_loss: 0.0012 - val_acc: 0.7664 Epoch 2570/3000 1712/1712 [==============================] - 1s - loss: 8.8708e-04 - acc: 0.8400 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 2571/3000 1712/1712 [==============================] - 1s - loss: 7.5592e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 2572/3000 1712/1712 [==============================] - 1s - loss: 7.9402e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 2573/3000 1712/1712 [==============================] - 1s - loss: 8.4002e-04 - acc: 0.8300 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 2574/3000 1712/1712 [==============================] - 1s - loss: 8.0966e-04 - acc: 0.8254 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 2575/3000 1712/1712 [==============================] - 1s - loss: 7.7499e-04 - acc: 0.8359 - val_loss: 0.0014 - val_acc: 0.7944 Epoch 2576/3000 1712/1712 [==============================] - 1s - loss: 8.2598e-04 - acc: 0.8248 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 2577/3000 1712/1712 [==============================] - 1s - loss: 8.0215e-04 - acc: 0.8347 - val_loss: 0.0023 - val_acc: 0.7593 Epoch 2578/3000 1712/1712 [==============================] - 1s - loss: 7.8119e-04 - acc: 0.8411 - val_loss: 0.0021 - val_acc: 0.7196 Epoch 2579/3000 1712/1712 [==============================] - 1s - loss: 8.9017e-04 - acc: 0.8201 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 2580/3000 1712/1712 [==============================] - 1s - loss: 7.8723e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 2581/3000 1712/1712 [==============================] - 1s - loss: 8.7761e-04 - acc: 0.8318 - val_loss: 0.0023 - val_acc: 0.7430 Epoch 2582/3000 1712/1712 [==============================] - 1s - loss: 7.6318e-04 - acc: 0.8411 - val_loss: 0.0014 - val_acc: 0.7664 Epoch 2583/3000 1712/1712 [==============================] - 1s - loss: 8.1831e-04 - acc: 0.8254 - val_loss: 0.0011 - val_acc: 0.8061 Epoch 2584/3000 1712/1712 [==============================] - 1s - loss: 8.2097e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2585/3000 1712/1712 [==============================] - 1s - loss: 7.4485e-04 - acc: 0.8493 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 2586/3000 1712/1712 [==============================] - 1s - loss: 8.4347e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2587/3000 1712/1712 [==============================] - 1s - loss: 9.0542e-04 - acc: 0.8388 - val_loss: 0.0015 - val_acc: 0.7336 Epoch 2588/3000 1712/1712 [==============================] - 1s - loss: 7.8755e-04 - acc: 0.8370 - val_loss: 0.0016 - val_acc: 0.7780 Epoch 2589/3000 1712/1712 [==============================] - 1s - loss: 7.9816e-04 - acc: 0.8429 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 2590/3000 1712/1712 [==============================] - 1s - loss: 8.7079e-04 - acc: 0.8259 - val_loss: 0.0016 - val_acc: 0.7734 Epoch 2591/3000 1712/1712 [==============================] - 1s - loss: 6.7983e-04 - acc: 0.8575 - val_loss: 0.0028 - val_acc: 0.7009 Epoch 2592/3000 1712/1712 [==============================] - 1s - loss: 8.5247e-04 - acc: 0.8236 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 2593/3000 1712/1712 [==============================] - 1s - loss: 8.0504e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2594/3000 1712/1712 [==============================] - 1s - loss: 7.0772e-04 - acc: 0.8487 - val_loss: 0.0016 - val_acc: 0.7757 Epoch 2595/3000 1712/1712 [==============================] - 1s - loss: 8.6156e-04 - acc: 0.8207 - val_loss: 0.0018 - val_acc: 0.7710 Epoch 2596/3000 1712/1712 [==============================] - 1s - loss: 8.5956e-04 - acc: 0.8324 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 2597/3000 1712/1712 [==============================] - 1s - loss: 7.7328e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2598/3000 1712/1712 [==============================] - 1s - loss: 7.6307e-04 - acc: 0.8353 - val_loss: 0.0024 - val_acc: 0.7220 Epoch 2599/3000 1712/1712 [==============================] - 1s - loss: 8.6564e-04 - acc: 0.8131 - val_loss: 0.0020 - val_acc: 0.7570 Epoch 2600/3000 1712/1712 [==============================] - 1s - loss: 7.7411e-04 - acc: 0.8312 - val_loss: 0.0020 - val_acc: 0.7313 Epoch 2601/3000 1712/1712 [==============================] - 1s - loss: 7.5101e-04 - acc: 0.8429 - val_loss: 0.0016 - val_acc: 0.7640 Epoch 2602/3000 1712/1712 [==============================] - 1s - loss: 8.2278e-04 - acc: 0.8528 - val_loss: 0.0018 - val_acc: 0.7500 Epoch 2603/3000 1712/1712 [==============================] - 1s - loss: 8.7210e-04 - acc: 0.8265 - val_loss: 0.0017 - val_acc: 0.7593 Epoch 2604/3000 1712/1712 [==============================] - 1s - loss: 7.4432e-04 - acc: 0.8283 - val_loss: 0.0019 - val_acc: 0.7313 Epoch 2605/3000 1712/1712 [==============================] - 1s - loss: 8.2667e-04 - acc: 0.8364 - val_loss: 0.0017 - val_acc: 0.7710 Epoch 2606/3000 1712/1712 [==============================] - 1s - loss: 8.0376e-04 - acc: 0.8370 - val_loss: 0.0012 - val_acc: 0.7710 Epoch 2607/3000 1712/1712 [==============================] - 1s - loss: 8.0668e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 2608/3000 1712/1712 [==============================] - 1s - loss: 8.3387e-04 - acc: 0.8294 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 2609/3000 1712/1712 [==============================] - 1s - loss: 6.9451e-04 - acc: 0.8499 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 2610/3000 1712/1712 [==============================] - 1s - loss: 8.3020e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 2611/3000 1712/1712 [==============================] - 1s - loss: 7.6240e-04 - acc: 0.8230 - val_loss: 0.0019 - val_acc: 0.7570 Epoch 2612/3000 1712/1712 [==============================] - 1s - loss: 8.0391e-04 - acc: 0.8364 - val_loss: 0.0019 - val_acc: 0.7593 Epoch 2613/3000 1712/1712 [==============================] - 1s - loss: 7.7320e-04 - acc: 0.8364 - val_loss: 0.0022 - val_acc: 0.7079 Epoch 2614/3000 1712/1712 [==============================] - 1s - loss: 8.3439e-04 - acc: 0.8189 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 2615/3000 1712/1712 [==============================] - 1s - loss: 8.2927e-04 - acc: 0.8423 - val_loss: 0.0016 - val_acc: 0.7547 Epoch 2616/3000 1712/1712 [==============================] - 1s - loss: 7.7119e-04 - acc: 0.8452 - val_loss: 0.0016 - val_acc: 0.7617 Epoch 2617/3000 1712/1712 [==============================] - 1s - loss: 8.0285e-04 - acc: 0.8464 - val_loss: 0.0021 - val_acc: 0.7313 Epoch 2618/3000 1712/1712 [==============================] - 1s - loss: 8.0032e-04 - acc: 0.8493 - val_loss: 0.0019 - val_acc: 0.7780 Epoch 2619/3000 1712/1712 [==============================] - 1s - loss: 8.0323e-04 - acc: 0.8341 - val_loss: 0.0016 - val_acc: 0.7336 Epoch 2620/3000 1712/1712 [==============================] - 1s - loss: 8.0941e-04 - acc: 0.8125 - val_loss: 0.0025 - val_acc: 0.7220 Epoch 2621/3000 1712/1712 [==============================] - 1s - loss: 8.6793e-04 - acc: 0.8435 - val_loss: 0.0020 - val_acc: 0.7126 Epoch 2622/3000 1712/1712 [==============================] - 1s - loss: 8.0828e-04 - acc: 0.8341 - val_loss: 0.0019 - val_acc: 0.7150 Epoch 2623/3000 1712/1712 [==============================] - 1s - loss: 7.7359e-04 - acc: 0.8429 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 2624/3000 1712/1712 [==============================] - 1s - loss: 8.6884e-04 - acc: 0.8423 - val_loss: 0.0014 - val_acc: 0.7547 Epoch 2625/3000 1712/1712 [==============================] - 1s - loss: 7.1175e-04 - acc: 0.8318 - val_loss: 0.0023 - val_acc: 0.7407 Epoch 2626/3000 1712/1712 [==============================] - 1s - loss: 8.3098e-04 - acc: 0.8306 - val_loss: 0.0015 - val_acc: 0.7687 Epoch 2627/3000 1712/1712 [==============================] - 1s - loss: 8.9257e-04 - acc: 0.8172 - val_loss: 0.0016 - val_acc: 0.7640 Epoch 2628/3000 1712/1712 [==============================] - 1s - loss: 7.5578e-04 - acc: 0.8394 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 2629/3000 1712/1712 [==============================] - 1s - loss: 8.3790e-04 - acc: 0.8300 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 2630/3000 1712/1712 [==============================] - 1s - loss: 7.6305e-04 - acc: 0.8329 - val_loss: 0.0018 - val_acc: 0.7453 Epoch 2631/3000 1712/1712 [==============================] - 1s - loss: 8.0526e-04 - acc: 0.8359 - val_loss: 0.0019 - val_acc: 0.7220 Epoch 2632/3000 1712/1712 [==============================] - 1s - loss: 8.1872e-04 - acc: 0.8125 - val_loss: 0.0017 - val_acc: 0.7290 Epoch 2633/3000 1712/1712 [==============================] - 1s - loss: 7.9811e-04 - acc: 0.8277 - val_loss: 0.0017 - val_acc: 0.8037 Epoch 2634/3000 1712/1712 [==============================] - 1s - loss: 8.0786e-04 - acc: 0.8289 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 2635/3000 1712/1712 [==============================] - 1s - loss: 7.4718e-04 - acc: 0.8359 - val_loss: 0.0018 - val_acc: 0.7593 Epoch 2636/3000 1712/1712 [==============================] - 1s - loss: 7.7399e-04 - acc: 0.8324 - val_loss: 0.0020 - val_acc: 0.7079 Epoch 2637/3000 1712/1712 [==============================] - 1s - loss: 8.1770e-04 - acc: 0.8236 - val_loss: 0.0015 - val_acc: 0.7617 Epoch 2638/3000 1712/1712 [==============================] - 1s - loss: 8.2540e-04 - acc: 0.8417 - val_loss: 0.0019 - val_acc: 0.7593 Epoch 2639/3000 1712/1712 [==============================] - 1s - loss: 7.6192e-04 - acc: 0.8557 - val_loss: 0.0014 - val_acc: 0.7757 Epoch 2640/3000 1712/1712 [==============================] - 1s - loss: 8.1043e-04 - acc: 0.8405 - val_loss: 0.0012 - val_acc: 0.8271 Epoch 2641/3000 1712/1712 [==============================] - 1s - loss: 8.3098e-04 - acc: 0.8353 - val_loss: 0.0021 - val_acc: 0.7547 Epoch 2642/3000 1712/1712 [==============================] - 1s - loss: 7.2357e-04 - acc: 0.8394 - val_loss: 0.0018 - val_acc: 0.7290 Epoch 2643/3000 1712/1712 [==============================] - 1s - loss: 8.8500e-04 - acc: 0.8189 - val_loss: 0.0017 - val_acc: 0.7196 Epoch 2644/3000 1712/1712 [==============================] - 1s - loss: 7.9216e-04 - acc: 0.8306 - val_loss: 0.0017 - val_acc: 0.7593 Epoch 2645/3000 1712/1712 [==============================] - 1s - loss: 7.8558e-04 - acc: 0.8370 - val_loss: 0.0015 - val_acc: 0.7874 Epoch 2646/3000 1712/1712 [==============================] - 1s - loss: 8.0108e-04 - acc: 0.8417 - val_loss: 0.0018 - val_acc: 0.7336 Epoch 2647/3000 1712/1712 [==============================] - 1s - loss: 7.4609e-04 - acc: 0.8446 - val_loss: 0.0016 - val_acc: 0.7827 Epoch 2648/3000 1712/1712 [==============================] - 1s - loss: 8.1248e-04 - acc: 0.8289 - val_loss: 0.0018 - val_acc: 0.7313 Epoch 2649/3000 1712/1712 [==============================] - 1s - loss: 7.9417e-04 - acc: 0.8324 - val_loss: 0.0017 - val_acc: 0.7780 Epoch 2650/3000 1712/1712 [==============================] - 1s - loss: 8.0997e-04 - acc: 0.8429 - val_loss: 0.0021 - val_acc: 0.7126 Epoch 2651/3000 1712/1712 [==============================] - 1s - loss: 8.1892e-04 - acc: 0.8452 - val_loss: 0.0016 - val_acc: 0.7687 Epoch 2652/3000 1712/1712 [==============================] - 1s - loss: 8.0695e-04 - acc: 0.8201 - val_loss: 0.0019 - val_acc: 0.7430 Epoch 2653/3000 1712/1712 [==============================] - 1s - loss: 8.0022e-04 - acc: 0.8458 - val_loss: 0.0019 - val_acc: 0.7477 Epoch 2654/3000 1712/1712 [==============================] - 1s - loss: 7.8851e-04 - acc: 0.8400 - val_loss: 0.0018 - val_acc: 0.7383 Epoch 2655/3000 1712/1712 [==============================] - 1s - loss: 8.2702e-04 - acc: 0.8318 - val_loss: 0.0017 - val_acc: 0.7664 Epoch 2656/3000 1712/1712 [==============================] - 1s - loss: 8.1131e-04 - acc: 0.8499 - val_loss: 0.0018 - val_acc: 0.7547 Epoch 2657/3000 1712/1712 [==============================] - 1s - loss: 7.9083e-04 - acc: 0.8353 - val_loss: 0.0018 - val_acc: 0.7430 Epoch 2658/3000 1712/1712 [==============================] - 1s - loss: 8.2606e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 2659/3000 1712/1712 [==============================] - 1s - loss: 8.8528e-04 - acc: 0.8458 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 2660/3000 1712/1712 [==============================] - 1s - loss: 7.6855e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 2661/3000 1712/1712 [==============================] - 1s - loss: 7.9760e-04 - acc: 0.8405 - val_loss: 0.0019 - val_acc: 0.7290 Epoch 2662/3000 1712/1712 [==============================] - 1s - loss: 8.1714e-04 - acc: 0.8440 - val_loss: 0.0015 - val_acc: 0.7500 Epoch 2663/3000 1712/1712 [==============================] - 1s - loss: 7.5688e-04 - acc: 0.8394 - val_loss: 0.0023 - val_acc: 0.7220 Epoch 2664/3000 1712/1712 [==============================] - 1s - loss: 8.6843e-04 - acc: 0.8417 - val_loss: 0.0020 - val_acc: 0.7313 Epoch 2665/3000 1712/1712 [==============================] - 1s - loss: 7.7079e-04 - acc: 0.8341 - val_loss: 0.0019 - val_acc: 0.7407 Epoch 2666/3000 1712/1712 [==============================] - 1s - loss: 8.2844e-04 - acc: 0.8294 - val_loss: 0.0014 - val_acc: 0.8131 Epoch 2667/3000 1712/1712 [==============================] - 1s - loss: 7.9590e-04 - acc: 0.8294 - val_loss: 0.0014 - val_acc: 0.7804 Epoch 2668/3000 1712/1712 [==============================] - 1s - loss: 8.1521e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2669/3000 1712/1712 [==============================] - 1s - loss: 7.8550e-04 - acc: 0.8376 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 2670/3000 1712/1712 [==============================] - 1s - loss: 8.5624e-04 - acc: 0.8306 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 2671/3000 1712/1712 [==============================] - 1s - loss: 7.9539e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 2672/3000 1712/1712 [==============================] - 1s - loss: 7.6093e-04 - acc: 0.8318 - val_loss: 0.0014 - val_acc: 0.7593 Epoch 2673/3000 1712/1712 [==============================] - 1s - loss: 8.2020e-04 - acc: 0.8189 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 2674/3000 1712/1712 [==============================] - 1s - loss: 7.6525e-04 - acc: 0.8435 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2675/3000 1712/1712 [==============================] - 1s - loss: 7.7845e-04 - acc: 0.8335 - val_loss: 0.0013 - val_acc: 0.7874 Epoch 2676/3000 1712/1712 [==============================] - 1s - loss: 8.1585e-04 - acc: 0.8347 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 2677/3000 1712/1712 [==============================] - 1s - loss: 8.0572e-04 - acc: 0.8172 - val_loss: 0.0014 - val_acc: 0.7617 Epoch 2678/3000 1712/1712 [==============================] - 1s - loss: 7.8090e-04 - acc: 0.8259 - val_loss: 0.0016 - val_acc: 0.7850 Epoch 2679/3000 1712/1712 [==============================] - 1s - loss: 7.1346e-04 - acc: 0.8493 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 2680/3000 1712/1712 [==============================] - 1s - loss: 8.3042e-04 - acc: 0.8207 - val_loss: 0.0017 - val_acc: 0.7874 Epoch 2681/3000 1712/1712 [==============================] - 1s - loss: 8.0532e-04 - acc: 0.8335 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 2682/3000 1712/1712 [==============================] - 1s - loss: 8.4983e-04 - acc: 0.8353 - val_loss: 0.0018 - val_acc: 0.7150 Epoch 2683/3000 1712/1712 [==============================] - 1s - loss: 8.1959e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 2684/3000 1712/1712 [==============================] - 1s - loss: 7.6546e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2685/3000 1712/1712 [==============================] - 1s - loss: 8.4464e-04 - acc: 0.8166 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 2686/3000 1712/1712 [==============================] - 1s - loss: 7.8866e-04 - acc: 0.8411 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 2687/3000 1712/1712 [==============================] - 1s - loss: 8.3774e-04 - acc: 0.8376 - val_loss: 0.0014 - val_acc: 0.7921 Epoch 2688/3000 1712/1712 [==============================] - 1s - loss: 7.7398e-04 - acc: 0.8446 - val_loss: 0.0014 - val_acc: 0.7944 Epoch 2689/3000 1712/1712 [==============================] - 1s - loss: 7.2203e-04 - acc: 0.8487 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 2690/3000 1712/1712 [==============================] - 1s - loss: 8.4731e-04 - acc: 0.8446 - val_loss: 0.0013 - val_acc: 0.8061 Epoch 2691/3000 1712/1712 [==============================] - 1s - loss: 7.7848e-04 - acc: 0.8382 - val_loss: 0.0014 - val_acc: 0.7921 Epoch 2692/3000 1712/1712 [==============================] - 1s - loss: 8.3673e-04 - acc: 0.8312 - val_loss: 0.0018 - val_acc: 0.7243 Epoch 2693/3000 1712/1712 [==============================] - 1s - loss: 7.6935e-04 - acc: 0.8405 - val_loss: 0.0018 - val_acc: 0.7407 Epoch 2694/3000 1712/1712 [==============================] - 1s - loss: 7.4236e-04 - acc: 0.8475 - val_loss: 0.0016 - val_acc: 0.7570 Epoch 2695/3000 1712/1712 [==============================] - 1s - loss: 8.2714e-04 - acc: 0.8329 - val_loss: 0.0026 - val_acc: 0.7360 Epoch 2696/3000 1712/1712 [==============================] - 1s - loss: 8.7480e-04 - acc: 0.8446 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 2697/3000 1712/1712 [==============================] - 1s - loss: 7.3998e-04 - acc: 0.8405 - val_loss: 0.0021 - val_acc: 0.7150 Epoch 2698/3000 1712/1712 [==============================] - 1s - loss: 8.0249e-04 - acc: 0.8364 - val_loss: 0.0018 - val_acc: 0.7150 Epoch 2699/3000 1712/1712 [==============================] - 1s - loss: 7.9429e-04 - acc: 0.8277 - val_loss: 0.0015 - val_acc: 0.7266 Epoch 2700/3000 1712/1712 [==============================] - 1s - loss: 8.4214e-04 - acc: 0.8283 - val_loss: 0.0014 - val_acc: 0.7593 Epoch 2701/3000 1712/1712 [==============================] - 1s - loss: 7.7079e-04 - acc: 0.8423 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2702/3000 1712/1712 [==============================] - 1s - loss: 7.5699e-04 - acc: 0.8394 - val_loss: 0.0020 - val_acc: 0.7477 Epoch 2703/3000 1712/1712 [==============================] - 1s - loss: 8.3698e-04 - acc: 0.8464 - val_loss: 0.0017 - val_acc: 0.7383 Epoch 2704/3000 1712/1712 [==============================] - 1s - loss: 8.0467e-04 - acc: 0.8423 - val_loss: 0.0020 - val_acc: 0.7103 Epoch 2705/3000 1712/1712 [==============================] - 1s - loss: 8.1804e-04 - acc: 0.8329 - val_loss: 0.0019 - val_acc: 0.7874 Epoch 2706/3000 1712/1712 [==============================] - 1s - loss: 8.4925e-04 - acc: 0.8359 - val_loss: 0.0017 - val_acc: 0.7360 Epoch 2707/3000 1712/1712 [==============================] - 1s - loss: 7.2895e-04 - acc: 0.8370 - val_loss: 0.0022 - val_acc: 0.7196 Epoch 2708/3000 1712/1712 [==============================] - 1s - loss: 8.0513e-04 - acc: 0.8218 - val_loss: 0.0016 - val_acc: 0.7290 Epoch 2709/3000 1712/1712 [==============================] - 1s - loss: 8.0221e-04 - acc: 0.8306 - val_loss: 0.0016 - val_acc: 0.7547 Epoch 2710/3000 1712/1712 [==============================] - 1s - loss: 7.8484e-04 - acc: 0.8283 - val_loss: 0.0020 - val_acc: 0.7547 Epoch 2711/3000 1712/1712 [==============================] - 1s - loss: 8.1585e-04 - acc: 0.8394 - val_loss: 0.0020 - val_acc: 0.7243 Epoch 2712/3000 1712/1712 [==============================] - 1s - loss: 8.3356e-04 - acc: 0.8405 - val_loss: 0.0018 - val_acc: 0.7687 Epoch 2713/3000 1712/1712 [==============================] - 1s - loss: 8.2460e-04 - acc: 0.8405 - val_loss: 0.0022 - val_acc: 0.7173 Epoch 2714/3000 1712/1712 [==============================] - 1s - loss: 7.5173e-04 - acc: 0.8470 - val_loss: 0.0019 - val_acc: 0.7196 Epoch 2715/3000 1712/1712 [==============================] - 1s - loss: 8.8138e-04 - acc: 0.8271 - val_loss: 0.0018 - val_acc: 0.7570 Epoch 2716/3000 1712/1712 [==============================] - 1s - loss: 7.8007e-04 - acc: 0.8382 - val_loss: 0.0018 - val_acc: 0.7780 Epoch 2717/3000 1712/1712 [==============================] - 1s - loss: 7.8641e-04 - acc: 0.8388 - val_loss: 0.0018 - val_acc: 0.7570 Epoch 2718/3000 1712/1712 [==============================] - 1s - loss: 7.8724e-04 - acc: 0.8394 - val_loss: 0.0023 - val_acc: 0.6986 Epoch 2719/3000 1712/1712 [==============================] - 1s - loss: 8.0036e-04 - acc: 0.8265 - val_loss: 0.0019 - val_acc: 0.7570 Epoch 2720/3000 1712/1712 [==============================] - 1s - loss: 7.7225e-04 - acc: 0.8464 - val_loss: 0.0018 - val_acc: 0.7196 Epoch 2721/3000 1712/1712 [==============================] - 1s - loss: 7.7370e-04 - acc: 0.8353 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2722/3000 1712/1712 [==============================] - 1s - loss: 8.1023e-04 - acc: 0.8382 - val_loss: 0.0014 - val_acc: 0.8037 Epoch 2723/3000 1712/1712 [==============================] - 1s - loss: 7.7597e-04 - acc: 0.8382 - val_loss: 0.0012 - val_acc: 0.7991 Epoch 2724/3000 1712/1712 [==============================] - 1s - loss: 7.9958e-04 - acc: 0.8376 - val_loss: 0.0014 - val_acc: 0.7640 Epoch 2725/3000 1712/1712 [==============================] - 1s - loss: 7.7341e-04 - acc: 0.8400 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 2726/3000 1712/1712 [==============================] - 1s - loss: 8.2321e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2727/3000 1712/1712 [==============================] - 1s - loss: 7.7499e-04 - acc: 0.8452 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 2728/3000 1712/1712 [==============================] - 1s - loss: 8.0921e-04 - acc: 0.8335 - val_loss: 0.0018 - val_acc: 0.7593 Epoch 2729/3000 1712/1712 [==============================] - 1s - loss: 8.2831e-04 - acc: 0.8329 - val_loss: 0.0016 - val_acc: 0.7547 Epoch 2730/3000 1712/1712 [==============================] - 1s - loss: 7.4979e-04 - acc: 0.8347 - val_loss: 0.0020 - val_acc: 0.7383 Epoch 2731/3000 1712/1712 [==============================] - 1s - loss: 8.8336e-04 - acc: 0.8306 - val_loss: 0.0017 - val_acc: 0.7944 Epoch 2732/3000 1712/1712 [==============================] - 1s - loss: 7.8063e-04 - acc: 0.8446 - val_loss: 0.0018 - val_acc: 0.7126 Epoch 2733/3000 1712/1712 [==============================] - 1s - loss: 7.6066e-04 - acc: 0.8318 - val_loss: 0.0021 - val_acc: 0.7407 Epoch 2734/3000 1712/1712 [==============================] - 1s - loss: 8.5456e-04 - acc: 0.8359 - val_loss: 0.0020 - val_acc: 0.7477 Epoch 2735/3000 1712/1712 [==============================] - 1s - loss: 8.4357e-04 - acc: 0.8300 - val_loss: 0.0015 - val_acc: 0.7593 Epoch 2736/3000 1712/1712 [==============================] - 1s - loss: 7.8514e-04 - acc: 0.8429 - val_loss: 0.0016 - val_acc: 0.7500 Epoch 2737/3000 1712/1712 [==============================] - 1s - loss: 8.5876e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 2738/3000 1712/1712 [==============================] - 1s - loss: 7.8531e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2739/3000 1712/1712 [==============================] - 1s - loss: 7.7073e-04 - acc: 0.8423 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 2740/3000 1712/1712 [==============================] - 1s - loss: 8.4929e-04 - acc: 0.8329 - val_loss: 0.0013 - val_acc: 0.8107 Epoch 2741/3000 1712/1712 [==============================] - 1s - loss: 7.8078e-04 - acc: 0.8294 - val_loss: 0.0012 - val_acc: 0.7827 Epoch 2742/3000 1712/1712 [==============================] - 1s - loss: 8.4024e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7734 Epoch 2743/3000 1712/1712 [==============================] - 1s - loss: 7.6644e-04 - acc: 0.8312 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2744/3000 1712/1712 [==============================] - 1s - loss: 8.3220e-04 - acc: 0.8353 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2745/3000 1712/1712 [==============================] - 1s - loss: 7.5760e-04 - acc: 0.8242 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 2746/3000 1712/1712 [==============================] - 1s - loss: 9.1870e-04 - acc: 0.8359 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 2747/3000 1712/1712 [==============================] - 1s - loss: 7.1383e-04 - acc: 0.8440 - val_loss: 0.0012 - val_acc: 0.8178 Epoch 2748/3000 1712/1712 [==============================] - 1s - loss: 8.3401e-04 - acc: 0.8382 - val_loss: 0.0016 - val_acc: 0.7336 Epoch 2749/3000 1712/1712 [==============================] - 1s - loss: 8.2222e-04 - acc: 0.8324 - val_loss: 0.0024 - val_acc: 0.7150 Epoch 2750/3000 1712/1712 [==============================] - 1s - loss: 8.5986e-04 - acc: 0.8429 - val_loss: 0.0019 - val_acc: 0.7430 Epoch 2751/3000 1712/1712 [==============================] - 1s - loss: 7.4779e-04 - acc: 0.8329 - val_loss: 0.0022 - val_acc: 0.7547 Epoch 2752/3000 1712/1712 [==============================] - 1s - loss: 7.7823e-04 - acc: 0.8458 - val_loss: 0.0021 - val_acc: 0.7126 Epoch 2753/3000 1712/1712 [==============================] - 1s - loss: 8.2461e-04 - acc: 0.8364 - val_loss: 0.0019 - val_acc: 0.7196 Epoch 2754/3000 1712/1712 [==============================] - 1s - loss: 7.6596e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 2755/3000 1712/1712 [==============================] - 1s - loss: 8.2474e-04 - acc: 0.8446 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2756/3000 1712/1712 [==============================] - 1s - loss: 7.6072e-04 - acc: 0.8411 - val_loss: 0.0012 - val_acc: 0.7734 Epoch 2757/3000 1712/1712 [==============================] - 1s - loss: 8.1746e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7640 Epoch 2758/3000 1712/1712 [==============================] - 1s - loss: 8.1005e-04 - acc: 0.8388 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 2759/3000 1712/1712 [==============================] - 1s - loss: 8.6709e-04 - acc: 0.8411 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 2760/3000 1712/1712 [==============================] - 1s - loss: 7.1423e-04 - acc: 0.8464 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 2761/3000 1712/1712 [==============================] - 1s - loss: 7.8802e-04 - acc: 0.8400 - val_loss: 0.0012 - val_acc: 0.7850 Epoch 2762/3000 1712/1712 [==============================] - 1s - loss: 9.0489e-04 - acc: 0.8172 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 2763/3000 1712/1712 [==============================] - 1s - loss: 8.3223e-04 - acc: 0.8411 - val_loss: 0.0019 - val_acc: 0.7593 Epoch 2764/3000 1712/1712 [==============================] - 1s - loss: 7.9108e-04 - acc: 0.8376 - val_loss: 0.0017 - val_acc: 0.7430 Epoch 2765/3000 1712/1712 [==============================] - 1s - loss: 8.1406e-04 - acc: 0.8370 - val_loss: 0.0015 - val_acc: 0.7734 Epoch 2766/3000 1712/1712 [==============================] - 1s - loss: 8.2240e-04 - acc: 0.8364 - val_loss: 0.0014 - val_acc: 0.7500 Epoch 2767/3000 1712/1712 [==============================] - 1s - loss: 7.4340e-04 - acc: 0.8470 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2768/3000 1712/1712 [==============================] - 1s - loss: 7.9761e-04 - acc: 0.8394 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 2769/3000 1712/1712 [==============================] - 1s - loss: 8.1415e-04 - acc: 0.8183 - val_loss: 0.0015 - val_acc: 0.7477 Epoch 2770/3000 1712/1712 [==============================] - 1s - loss: 7.6985e-04 - acc: 0.8417 - val_loss: 0.0018 - val_acc: 0.7570 Epoch 2771/3000 1712/1712 [==============================] - 1s - loss: 8.2400e-04 - acc: 0.8271 - val_loss: 0.0016 - val_acc: 0.7874 Epoch 2772/3000 1712/1712 [==============================] - 1s - loss: 8.0638e-04 - acc: 0.8376 - val_loss: 0.0022 - val_acc: 0.7056 Epoch 2773/3000 1712/1712 [==============================] - 1s - loss: 7.3854e-04 - acc: 0.8394 - val_loss: 0.0012 - val_acc: 0.7757 Epoch 2774/3000 1712/1712 [==============================] - 1s - loss: 8.7816e-04 - acc: 0.8429 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2775/3000 1712/1712 [==============================] - 1s - loss: 7.5256e-04 - acc: 0.8370 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2776/3000 1712/1712 [==============================] - 1s - loss: 7.9996e-04 - acc: 0.8470 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 2777/3000 1712/1712 [==============================] - 1s - loss: 7.6624e-04 - acc: 0.8435 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 2778/3000 1712/1712 [==============================] - 1s - loss: 8.2914e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7593 Epoch 2779/3000 1712/1712 [==============================] - 1s - loss: 8.5906e-04 - acc: 0.8178 - val_loss: 0.0014 - val_acc: 0.7944 Epoch 2780/3000 1712/1712 [==============================] - 1s - loss: 7.3267e-04 - acc: 0.8511 - val_loss: 0.0013 - val_acc: 0.7617 Epoch 2781/3000 1712/1712 [==============================] - 1s - loss: 7.5715e-04 - acc: 0.8364 - val_loss: 0.0019 - val_acc: 0.7664 Epoch 2782/3000 1712/1712 [==============================] - 1s - loss: 8.2590e-04 - acc: 0.8271 - val_loss: 0.0023 - val_acc: 0.7103 Epoch 2783/3000 1712/1712 [==============================] - 1s - loss: 7.6454e-04 - acc: 0.8405 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 2784/3000 1712/1712 [==============================] - 1s - loss: 9.4751e-04 - acc: 0.8265 - val_loss: 0.0014 - val_acc: 0.7477 Epoch 2785/3000 1712/1712 [==============================] - 1s - loss: 6.9795e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7780 Epoch 2786/3000 1712/1712 [==============================] - 1s - loss: 8.2785e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 2787/3000 1712/1712 [==============================] - 1s - loss: 8.2548e-04 - acc: 0.8183 - val_loss: 0.0015 - val_acc: 0.7780 Epoch 2788/3000 1712/1712 [==============================] - 1s - loss: 8.1370e-04 - acc: 0.8446 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2789/3000 1712/1712 [==============================] - 1s - loss: 8.1170e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2790/3000 1712/1712 [==============================] - 1s - loss: 7.9787e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2791/3000 1712/1712 [==============================] - 1s - loss: 8.4164e-04 - acc: 0.8364 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 2792/3000 1712/1712 [==============================] - 1s - loss: 7.3425e-04 - acc: 0.8516 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 2793/3000 1712/1712 [==============================] - 1s - loss: 8.7745e-04 - acc: 0.8195 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 2794/3000 1712/1712 [==============================] - 1s - loss: 7.7985e-04 - acc: 0.8417 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 2795/3000 1712/1712 [==============================] - 1s - loss: 8.0316e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 2796/3000 1712/1712 [==============================] - 1s - loss: 7.6308e-04 - acc: 0.8277 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 2797/3000 1712/1712 [==============================] - 1s - loss: 8.3790e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2798/3000 1712/1712 [==============================] - 1s - loss: 8.1050e-04 - acc: 0.8440 - val_loss: 0.0017 - val_acc: 0.7734 Epoch 2799/3000 1712/1712 [==============================] - 1s - loss: 7.7466e-04 - acc: 0.8312 - val_loss: 0.0020 - val_acc: 0.7547 Epoch 2800/3000 1712/1712 [==============================] - 1s - loss: 7.7558e-04 - acc: 0.8364 - val_loss: 0.0021 - val_acc: 0.7220 Epoch 2801/3000 1712/1712 [==============================] - 1s - loss: 8.7618e-04 - acc: 0.8306 - val_loss: 0.0016 - val_acc: 0.7640 Epoch 2802/3000 1712/1712 [==============================] - 1s - loss: 7.8626e-04 - acc: 0.8347 - val_loss: 0.0019 - val_acc: 0.7523 Epoch 2803/3000 1712/1712 [==============================] - 1s - loss: 8.1215e-04 - acc: 0.8341 - val_loss: 0.0015 - val_acc: 0.7477 Epoch 2804/3000 1712/1712 [==============================] - 1s - loss: 7.8748e-04 - acc: 0.8242 - val_loss: 0.0025 - val_acc: 0.7360 Epoch 2805/3000 1712/1712 [==============================] - 1s - loss: 8.6449e-04 - acc: 0.8248 - val_loss: 0.0015 - val_acc: 0.7407 Epoch 2806/3000 1712/1712 [==============================] - 1s - loss: 8.2959e-04 - acc: 0.8183 - val_loss: 0.0015 - val_acc: 0.7290 Epoch 2807/3000 1712/1712 [==============================] - 1s - loss: 7.9395e-04 - acc: 0.8335 - val_loss: 0.0020 - val_acc: 0.7290 Epoch 2808/3000 1712/1712 [==============================] - 1s - loss: 8.0791e-04 - acc: 0.8283 - val_loss: 0.0015 - val_acc: 0.7523 Epoch 2809/3000 1712/1712 [==============================] - 1s - loss: 8.6732e-04 - acc: 0.8283 - val_loss: 0.0020 - val_acc: 0.7150 Epoch 2810/3000 1712/1712 [==============================] - 1s - loss: 7.3316e-04 - acc: 0.8289 - val_loss: 0.0018 - val_acc: 0.7523 Epoch 2811/3000 1712/1712 [==============================] - 1s - loss: 8.8850e-04 - acc: 0.8248 - val_loss: 0.0021 - val_acc: 0.7173 Epoch 2812/3000 1712/1712 [==============================] - 1s - loss: 8.1662e-04 - acc: 0.8306 - val_loss: 0.0022 - val_acc: 0.7547 Epoch 2813/3000 1712/1712 [==============================] - 1s - loss: 8.1543e-04 - acc: 0.8201 - val_loss: 0.0020 - val_acc: 0.7617 Epoch 2814/3000 1712/1712 [==============================] - 1s - loss: 8.1695e-04 - acc: 0.8300 - val_loss: 0.0020 - val_acc: 0.7383 Epoch 2815/3000 1712/1712 [==============================] - 1s - loss: 7.7660e-04 - acc: 0.8335 - val_loss: 0.0018 - val_acc: 0.7243 Epoch 2816/3000 1712/1712 [==============================] - 1s - loss: 7.6704e-04 - acc: 0.8347 - val_loss: 0.0017 - val_acc: 0.8014 Epoch 2817/3000 1712/1712 [==============================] - 1s - loss: 8.1420e-04 - acc: 0.8254 - val_loss: 0.0023 - val_acc: 0.7430 Epoch 2818/3000 1712/1712 [==============================] - 1s - loss: 8.1093e-04 - acc: 0.8359 - val_loss: 0.0019 - val_acc: 0.7383 Epoch 2819/3000 1712/1712 [==============================] - 1s - loss: 7.5415e-04 - acc: 0.8242 - val_loss: 0.0021 - val_acc: 0.7407 Epoch 2820/3000 1712/1712 [==============================] - 1s - loss: 8.4422e-04 - acc: 0.8294 - val_loss: 0.0019 - val_acc: 0.7453 Epoch 2821/3000 1712/1712 [==============================] - 1s - loss: 8.0246e-04 - acc: 0.8236 - val_loss: 0.0016 - val_acc: 0.7850 Epoch 2822/3000 1712/1712 [==============================] - 1s - loss: 7.8371e-04 - acc: 0.8411 - val_loss: 0.0020 - val_acc: 0.7547 Epoch 2823/3000 1712/1712 [==============================] - 1s - loss: 8.4001e-04 - acc: 0.8400 - val_loss: 0.0015 - val_acc: 0.7874 Epoch 2824/3000 1712/1712 [==============================] - 1s - loss: 7.9182e-04 - acc: 0.8259 - val_loss: 0.0016 - val_acc: 0.7664 Epoch 2825/3000 1712/1712 [==============================] - 1s - loss: 8.0755e-04 - acc: 0.8341 - val_loss: 0.0015 - val_acc: 0.7921 Epoch 2826/3000 1712/1712 [==============================] - 1s - loss: 8.3681e-04 - acc: 0.8329 - val_loss: 0.0020 - val_acc: 0.7710 Epoch 2827/3000 1712/1712 [==============================] - 1s - loss: 7.3552e-04 - acc: 0.8423 - val_loss: 0.0013 - val_acc: 0.8061 Epoch 2828/3000 1712/1712 [==============================] - 1s - loss: 8.8983e-04 - acc: 0.8265 - val_loss: 0.0015 - val_acc: 0.8061 Epoch 2829/3000 1712/1712 [==============================] - 1s - loss: 8.2501e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 2830/3000 1712/1712 [==============================] - 1s - loss: 7.6124e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 2831/3000 1712/1712 [==============================] - 1s - loss: 8.4717e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 2832/3000 1712/1712 [==============================] - 1s - loss: 7.1697e-04 - acc: 0.8300 - val_loss: 0.0019 - val_acc: 0.7523 Epoch 2833/3000 1712/1712 [==============================] - 1s - loss: 8.1410e-04 - acc: 0.8359 - val_loss: 0.0017 - val_acc: 0.7850 Epoch 2834/3000 1712/1712 [==============================] - 1s - loss: 7.8752e-04 - acc: 0.8382 - val_loss: 0.0022 - val_acc: 0.7757 Epoch 2835/3000 1712/1712 [==============================] - 1s - loss: 9.0126e-04 - acc: 0.8283 - val_loss: 0.0017 - val_acc: 0.7313 Epoch 2836/3000 1712/1712 [==============================] - 1s - loss: 7.7750e-04 - acc: 0.8417 - val_loss: 0.0016 - val_acc: 0.7593 Epoch 2837/3000 1712/1712 [==============================] - 1s - loss: 7.8634e-04 - acc: 0.8306 - val_loss: 0.0021 - val_acc: 0.7617 Epoch 2838/3000 1712/1712 [==============================] - 1s - loss: 7.4007e-04 - acc: 0.8417 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 2839/3000 1712/1712 [==============================] - 1s - loss: 7.9265e-04 - acc: 0.8458 - val_loss: 0.0011 - val_acc: 0.8178 Epoch 2840/3000 1712/1712 [==============================] - 1s - loss: 7.8343e-04 - acc: 0.8452 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 2841/3000 1712/1712 [==============================] - 1s - loss: 8.4061e-04 - acc: 0.8405 - val_loss: 0.0017 - val_acc: 0.7383 Epoch 2842/3000 1712/1712 [==============================] - 1s - loss: 8.1510e-04 - acc: 0.8423 - val_loss: 0.0013 - val_acc: 0.7827 Epoch 2843/3000 1712/1712 [==============================] - 1s - loss: 8.1390e-04 - acc: 0.8306 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 2844/3000 1712/1712 [==============================] - 1s - loss: 7.9983e-04 - acc: 0.8283 - val_loss: 0.0012 - val_acc: 0.8061 Epoch 2845/3000 1712/1712 [==============================] - 1s - loss: 7.7407e-04 - acc: 0.8440 - val_loss: 0.0014 - val_acc: 0.7523 Epoch 2846/3000 1712/1712 [==============================] - 1s - loss: 7.7242e-04 - acc: 0.8335 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2847/3000 1712/1712 [==============================] - 1s - loss: 8.3171e-04 - acc: 0.8294 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 2848/3000 1712/1712 [==============================] - 1s - loss: 8.0246e-04 - acc: 0.8446 - val_loss: 0.0020 - val_acc: 0.7593 Epoch 2849/3000 1712/1712 [==============================] - 1s - loss: 7.6167e-04 - acc: 0.8312 - val_loss: 0.0020 - val_acc: 0.7453 Epoch 2850/3000 1712/1712 [==============================] - 1s - loss: 8.4718e-04 - acc: 0.8423 - val_loss: 0.0017 - val_acc: 0.7453 Epoch 2851/3000 1712/1712 [==============================] - 1s - loss: 7.9963e-04 - acc: 0.8324 - val_loss: 0.0016 - val_acc: 0.7921 Epoch 2852/3000 1712/1712 [==============================] - 1s - loss: 7.5653e-04 - acc: 0.8388 - val_loss: 0.0017 - val_acc: 0.7430 Epoch 2853/3000 1712/1712 [==============================] - 1s - loss: 8.2859e-04 - acc: 0.8359 - val_loss: 0.0020 - val_acc: 0.7243 Epoch 2854/3000 1712/1712 [==============================] - 1s - loss: 8.4418e-04 - acc: 0.8329 - val_loss: 0.0018 - val_acc: 0.7313 Epoch 2855/3000 1712/1712 [==============================] - 1s - loss: 7.7384e-04 - acc: 0.8458 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2856/3000 1712/1712 [==============================] - 1s - loss: 7.9930e-04 - acc: 0.8400 - val_loss: 0.0014 - val_acc: 0.7664 Epoch 2857/3000 1712/1712 [==============================] - 1s - loss: 8.0285e-04 - acc: 0.8382 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 2858/3000 1712/1712 [==============================] - 1s - loss: 8.2031e-04 - acc: 0.8487 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2859/3000 1712/1712 [==============================] - 1s - loss: 8.2526e-04 - acc: 0.8341 - val_loss: 0.0012 - val_acc: 0.7874 Epoch 2860/3000 1712/1712 [==============================] - 1s - loss: 7.2862e-04 - acc: 0.8388 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 2861/3000 1712/1712 [==============================] - 1s - loss: 7.9847e-04 - acc: 0.8265 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 2862/3000 1712/1712 [==============================] - 1s - loss: 8.4297e-04 - acc: 0.8242 - val_loss: 0.0011 - val_acc: 0.7944 Epoch 2863/3000 1712/1712 [==============================] - 1s - loss: 8.2116e-04 - acc: 0.8242 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 2864/3000 1712/1712 [==============================] - 1s - loss: 7.7977e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2865/3000 1712/1712 [==============================] - 1s - loss: 7.6817e-04 - acc: 0.8359 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 2866/3000 1712/1712 [==============================] - 1s - loss: 8.2834e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 2867/3000 1712/1712 [==============================] - 1s - loss: 8.0438e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 2868/3000 1712/1712 [==============================] - 1s - loss: 7.9046e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 2869/3000 1712/1712 [==============================] - 1s - loss: 7.9947e-04 - acc: 0.8581 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 2870/3000 1712/1712 [==============================] - 1s - loss: 7.7068e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 2871/3000 1712/1712 [==============================] - 1s - loss: 8.0505e-04 - acc: 0.8324 - val_loss: 0.0013 - val_acc: 0.7710 Epoch 2872/3000 1712/1712 [==============================] - 1s - loss: 7.9791e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7547 Epoch 2873/3000 1712/1712 [==============================] - 1s - loss: 8.9577e-04 - acc: 0.8248 - val_loss: 0.0023 - val_acc: 0.7033 Epoch 2874/3000 1712/1712 [==============================] - 1s - loss: 7.9126e-04 - acc: 0.8452 - val_loss: 0.0018 - val_acc: 0.7617 Epoch 2875/3000 1712/1712 [==============================] - 1s - loss: 8.4147e-04 - acc: 0.8364 - val_loss: 0.0015 - val_acc: 0.7991 Epoch 2876/3000 1712/1712 [==============================] - 1s - loss: 7.4197e-04 - acc: 0.8265 - val_loss: 0.0022 - val_acc: 0.7220 Epoch 2877/3000 1712/1712 [==============================] - 1s - loss: 7.8552e-04 - acc: 0.8312 - val_loss: 0.0016 - val_acc: 0.7593 Epoch 2878/3000 1712/1712 [==============================] - 1s - loss: 8.3017e-04 - acc: 0.8300 - val_loss: 0.0017 - val_acc: 0.7804 Epoch 2879/3000 1712/1712 [==============================] - 1s - loss: 7.7580e-04 - acc: 0.8394 - val_loss: 0.0019 - val_acc: 0.7173 Epoch 2880/3000 1712/1712 [==============================] - 1s - loss: 8.3760e-04 - acc: 0.8294 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 2881/3000 1712/1712 [==============================] - 1s - loss: 7.5448e-04 - acc: 0.8458 - val_loss: 0.0019 - val_acc: 0.7734 Epoch 2882/3000 1712/1712 [==============================] - 1s - loss: 8.1823e-04 - acc: 0.8370 - val_loss: 0.0017 - val_acc: 0.7804 Epoch 2883/3000 1712/1712 [==============================] - 1s - loss: 7.9653e-04 - acc: 0.8283 - val_loss: 0.0019 - val_acc: 0.7407 Epoch 2884/3000 1712/1712 [==============================] - 1s - loss: 8.4489e-04 - acc: 0.8324 - val_loss: 0.0018 - val_acc: 0.7664 Epoch 2885/3000 1712/1712 [==============================] - 1s - loss: 7.5682e-04 - acc: 0.8353 - val_loss: 0.0022 - val_acc: 0.7430 Epoch 2886/3000 1712/1712 [==============================] - 1s - loss: 8.5255e-04 - acc: 0.8400 - val_loss: 0.0016 - val_acc: 0.7593 Epoch 2887/3000 1712/1712 [==============================] - 1s - loss: 7.8275e-04 - acc: 0.8335 - val_loss: 0.0019 - val_acc: 0.7290 Epoch 2888/3000 1712/1712 [==============================] - 1s - loss: 7.8236e-04 - acc: 0.8429 - val_loss: 0.0019 - val_acc: 0.7173 Epoch 2889/3000 1712/1712 [==============================] - 1s - loss: 8.9564e-04 - acc: 0.8248 - val_loss: 0.0016 - val_acc: 0.7640 Epoch 2890/3000 1712/1712 [==============================] - 1s - loss: 7.7159e-04 - acc: 0.8347 - val_loss: 0.0023 - val_acc: 0.7126 Epoch 2891/3000 1712/1712 [==============================] - 1s - loss: 8.2850e-04 - acc: 0.8347 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 2892/3000 1712/1712 [==============================] - 1s - loss: 7.4883e-04 - acc: 0.8341 - val_loss: 0.0019 - val_acc: 0.7477 Epoch 2893/3000 1712/1712 [==============================] - 1s - loss: 8.3193e-04 - acc: 0.8364 - val_loss: 0.0019 - val_acc: 0.7687 Epoch 2894/3000 1712/1712 [==============================] - 1s - loss: 7.6499e-04 - acc: 0.8294 - val_loss: 0.0023 - val_acc: 0.7103 Epoch 2895/3000 1712/1712 [==============================] - 1s - loss: 8.2950e-04 - acc: 0.8218 - val_loss: 0.0017 - val_acc: 0.7313 Epoch 2896/3000 1712/1712 [==============================] - 1s - loss: 8.0167e-04 - acc: 0.8411 - val_loss: 0.0019 - val_acc: 0.7196 Epoch 2897/3000 1712/1712 [==============================] - 1s - loss: 7.9693e-04 - acc: 0.8329 - val_loss: 0.0016 - val_acc: 0.7921 Epoch 2898/3000 1712/1712 [==============================] - 1s - loss: 7.2120e-04 - acc: 0.8341 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2899/3000 1712/1712 [==============================] - 1s - loss: 8.8093e-04 - acc: 0.8353 - val_loss: 0.0014 - val_acc: 0.7897 Epoch 2900/3000 1712/1712 [==============================] - 1s - loss: 6.5113e-04 - acc: 0.8359 - val_loss: 0.0011 - val_acc: 0.8341 Epoch 2901/3000 1712/1712 [==============================] - 1s - loss: 9.2174e-04 - acc: 0.8242 - val_loss: 0.0018 - val_acc: 0.7383 Epoch 2902/3000 1712/1712 [==============================] - 1s - loss: 8.5341e-04 - acc: 0.8218 - val_loss: 0.0018 - val_acc: 0.7897 Epoch 2903/3000 1712/1712 [==============================] - 1s - loss: 7.5119e-04 - acc: 0.8405 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 2904/3000 1712/1712 [==============================] - 1s - loss: 8.7164e-04 - acc: 0.8218 - val_loss: 0.0016 - val_acc: 0.7664 Epoch 2905/3000 1712/1712 [==============================] - 1s - loss: 7.6071e-04 - acc: 0.8400 - val_loss: 0.0019 - val_acc: 0.7547 Epoch 2906/3000 1712/1712 [==============================] - 1s - loss: 7.6865e-04 - acc: 0.8254 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2907/3000 1712/1712 [==============================] - 1s - loss: 8.3194e-04 - acc: 0.8277 - val_loss: 0.0014 - val_acc: 0.7687 Epoch 2908/3000 1712/1712 [==============================] - 1s - loss: 7.6095e-04 - acc: 0.8353 - val_loss: 0.0016 - val_acc: 0.7477 Epoch 2909/3000 1712/1712 [==============================] - 1s - loss: 8.1453e-04 - acc: 0.8289 - val_loss: 0.0017 - val_acc: 0.7593 Epoch 2910/3000 1712/1712 [==============================] - 1s - loss: 8.2429e-04 - acc: 0.8435 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 2911/3000 1712/1712 [==============================] - 1s - loss: 7.5960e-04 - acc: 0.8388 - val_loss: 0.0024 - val_acc: 0.7220 Epoch 2912/3000 1712/1712 [==============================] - 1s - loss: 8.2032e-04 - acc: 0.8382 - val_loss: 0.0017 - val_acc: 0.7757 Epoch 2913/3000 1712/1712 [==============================] - 1s - loss: 7.9212e-04 - acc: 0.8283 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 2914/3000 1712/1712 [==============================] - 1s - loss: 9.0552e-04 - acc: 0.8137 - val_loss: 0.0013 - val_acc: 0.7967 Epoch 2915/3000 1712/1712 [==============================] - 1s - loss: 8.2583e-04 - acc: 0.8487 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 2916/3000 1712/1712 [==============================] - 1s - loss: 7.5308e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.8014 Epoch 2917/3000 1712/1712 [==============================] - 1s - loss: 8.4653e-04 - acc: 0.8289 - val_loss: 0.0012 - val_acc: 0.8201 Epoch 2918/3000 1712/1712 [==============================] - 1s - loss: 7.6614e-04 - acc: 0.8382 - val_loss: 0.0013 - val_acc: 0.7757 Epoch 2919/3000 1712/1712 [==============================] - 1s - loss: 7.8646e-04 - acc: 0.8481 - val_loss: 0.0014 - val_acc: 0.7640 Epoch 2920/3000 1712/1712 [==============================] - 1s - loss: 7.6956e-04 - acc: 0.8347 - val_loss: 0.0020 - val_acc: 0.7336 Epoch 2921/3000 1712/1712 [==============================] - 1s - loss: 8.0753e-04 - acc: 0.8259 - val_loss: 0.0019 - val_acc: 0.7547 Epoch 2922/3000 1712/1712 [==============================] - 1s - loss: 8.1815e-04 - acc: 0.8312 - val_loss: 0.0021 - val_acc: 0.7547 Epoch 2923/3000 1712/1712 [==============================] - 1s - loss: 7.9809e-04 - acc: 0.8499 - val_loss: 0.0017 - val_acc: 0.7664 Epoch 2924/3000 1712/1712 [==============================] - 1s - loss: 7.8990e-04 - acc: 0.8394 - val_loss: 0.0019 - val_acc: 0.7453 Epoch 2925/3000 1712/1712 [==============================] - 1s - loss: 7.8426e-04 - acc: 0.8335 - val_loss: 0.0020 - val_acc: 0.7150 Epoch 2926/3000 1712/1712 [==============================] - 1s - loss: 8.4934e-04 - acc: 0.8364 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 2927/3000 1712/1712 [==============================] - 1s - loss: 7.6015e-04 - acc: 0.8306 - val_loss: 0.0019 - val_acc: 0.7407 Epoch 2928/3000 1712/1712 [==============================] - 1s - loss: 8.8657e-04 - acc: 0.8254 - val_loss: 0.0019 - val_acc: 0.7290 Epoch 2929/3000 1712/1712 [==============================] - 1s - loss: 7.3386e-04 - acc: 0.8429 - val_loss: 0.0013 - val_acc: 0.8248 Epoch 2930/3000 1712/1712 [==============================] - 1s - loss: 8.4847e-04 - acc: 0.8207 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 2931/3000 1712/1712 [==============================] - 1s - loss: 7.8842e-04 - acc: 0.8417 - val_loss: 0.0014 - val_acc: 0.7780 Epoch 2932/3000 1712/1712 [==============================] - 1s - loss: 7.6801e-04 - acc: 0.8394 - val_loss: 0.0012 - val_acc: 0.8107 Epoch 2933/3000 1712/1712 [==============================] - 1s - loss: 8.0088e-04 - acc: 0.8388 - val_loss: 0.0014 - val_acc: 0.7921 Epoch 2934/3000 1712/1712 [==============================] - 1s - loss: 8.1215e-04 - acc: 0.8213 - val_loss: 0.0013 - val_acc: 0.7593 Epoch 2935/3000 1712/1712 [==============================] - 1s - loss: 7.9543e-04 - acc: 0.8370 - val_loss: 0.0013 - val_acc: 0.7921 Epoch 2936/3000 1712/1712 [==============================] - 1s - loss: 8.0262e-04 - acc: 0.8335 - val_loss: 0.0014 - val_acc: 0.7547 Epoch 2937/3000 1712/1712 [==============================] - 1s - loss: 8.2214e-04 - acc: 0.8318 - val_loss: 0.0014 - val_acc: 0.7967 Epoch 2938/3000 1712/1712 [==============================] - 1s - loss: 7.4090e-04 - acc: 0.8294 - val_loss: 0.0021 - val_acc: 0.7664 Epoch 2939/3000 1712/1712 [==============================] - 1s - loss: 8.2610e-04 - acc: 0.8417 - val_loss: 0.0015 - val_acc: 0.7477 Epoch 2940/3000 1712/1712 [==============================] - 1s - loss: 7.8721e-04 - acc: 0.8376 - val_loss: 0.0018 - val_acc: 0.7477 Epoch 2941/3000 1712/1712 [==============================] - 1s - loss: 8.0236e-04 - acc: 0.8411 - val_loss: 0.0015 - val_acc: 0.7944 Epoch 2942/3000 1712/1712 [==============================] - 1s - loss: 7.8237e-04 - acc: 0.8458 - val_loss: 0.0017 - val_acc: 0.7780 Epoch 2943/3000 1712/1712 [==============================] - 1s - loss: 7.9512e-04 - acc: 0.8294 - val_loss: 0.0017 - val_acc: 0.7430 Epoch 2944/3000 1712/1712 [==============================] - 1s - loss: 8.1244e-04 - acc: 0.8411 - val_loss: 0.0017 - val_acc: 0.7290 Epoch 2945/3000 1712/1712 [==============================] - 1s - loss: 8.4984e-04 - acc: 0.8376 - val_loss: 0.0018 - val_acc: 0.7593 Epoch 2946/3000 1712/1712 [==============================] - 1s - loss: 7.3432e-04 - acc: 0.8405 - val_loss: 0.0022 - val_acc: 0.7453 Epoch 2947/3000 1712/1712 [==============================] - 1s - loss: 8.6718e-04 - acc: 0.8318 - val_loss: 0.0023 - val_acc: 0.7033 Epoch 2948/3000 1712/1712 [==============================] - 1s - loss: 7.9388e-04 - acc: 0.8230 - val_loss: 0.0011 - val_acc: 0.7850 Epoch 2949/3000 1712/1712 [==============================] - 1s - loss: 8.2844e-04 - acc: 0.8423 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 2950/3000 1712/1712 [==============================] - 1s - loss: 8.0216e-04 - acc: 0.8364 - val_loss: 0.0013 - val_acc: 0.8084 Epoch 2951/3000 1712/1712 [==============================] - 1s - loss: 7.8115e-04 - acc: 0.8400 - val_loss: 0.0013 - val_acc: 0.7991 Epoch 2952/3000 1712/1712 [==============================] - 1s - loss: 8.0299e-04 - acc: 0.8353 - val_loss: 0.0013 - val_acc: 0.7897 Epoch 2953/3000 1712/1712 [==============================] - 1s - loss: 8.8018e-04 - acc: 0.8376 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 2954/3000 1712/1712 [==============================] - 1s - loss: 7.6689e-04 - acc: 0.8359 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2955/3000 1712/1712 [==============================] - 1s - loss: 7.6393e-04 - acc: 0.8353 - val_loss: 0.0013 - val_acc: 0.7664 Epoch 2956/3000 1712/1712 [==============================] - 1s - loss: 8.3456e-04 - acc: 0.8318 - val_loss: 0.0013 - val_acc: 0.7780 Epoch 2957/3000 1712/1712 [==============================] - 1s - loss: 8.0998e-04 - acc: 0.8242 - val_loss: 0.0013 - val_acc: 0.7570 Epoch 2958/3000 1712/1712 [==============================] - 1s - loss: 8.3119e-04 - acc: 0.8329 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2959/3000 1712/1712 [==============================] - 1s - loss: 7.5716e-04 - acc: 0.8435 - val_loss: 0.0012 - val_acc: 0.7944 Epoch 2960/3000 1712/1712 [==============================] - 1s - loss: 8.3224e-04 - acc: 0.8335 - val_loss: 0.0012 - val_acc: 0.7921 Epoch 2961/3000 1712/1712 [==============================] - 1s - loss: 7.9905e-04 - acc: 0.8312 - val_loss: 0.0012 - val_acc: 0.7897 Epoch 2962/3000 1712/1712 [==============================] - 1s - loss: 8.1290e-04 - acc: 0.8435 - val_loss: 0.0012 - val_acc: 0.8037 Epoch 2963/3000 1712/1712 [==============================] - 1s - loss: 6.9907e-04 - acc: 0.8376 - val_loss: 0.0026 - val_acc: 0.7103 Epoch 2964/3000 1712/1712 [==============================] - 1s - loss: 8.7866e-04 - acc: 0.8289 - val_loss: 0.0015 - val_acc: 0.7850 Epoch 2965/3000 1712/1712 [==============================] - 1s - loss: 7.6844e-04 - acc: 0.8353 - val_loss: 0.0015 - val_acc: 0.7617 Epoch 2966/3000 1712/1712 [==============================] - 1s - loss: 8.1044e-04 - acc: 0.8364 - val_loss: 0.0013 - val_acc: 0.7687 Epoch 2967/3000 1712/1712 [==============================] - 1s - loss: 7.6786e-04 - acc: 0.8265 - val_loss: 0.0014 - val_acc: 0.8061 Epoch 2968/3000 1712/1712 [==============================] - 1s - loss: 8.3798e-04 - acc: 0.8400 - val_loss: 0.0015 - val_acc: 0.7757 Epoch 2969/3000 1712/1712 [==============================] - 1s - loss: 8.0547e-04 - acc: 0.8400 - val_loss: 0.0016 - val_acc: 0.7477 Epoch 2970/3000 1712/1712 [==============================] - 1s - loss: 7.6052e-04 - acc: 0.8353 - val_loss: 0.0016 - val_acc: 0.7383 Epoch 2971/3000 1712/1712 [==============================] - 1s - loss: 7.7913e-04 - acc: 0.8423 - val_loss: 0.0017 - val_acc: 0.7477 Epoch 2972/3000 1712/1712 [==============================] - 1s - loss: 8.4141e-04 - acc: 0.8382 - val_loss: 0.0018 - val_acc: 0.7383 Epoch 2973/3000 1712/1712 [==============================] - 1s - loss: 7.6111e-04 - acc: 0.8341 - val_loss: 0.0018 - val_acc: 0.7220 Epoch 2974/3000 1712/1712 [==============================] - 1s - loss: 8.6043e-04 - acc: 0.8224 - val_loss: 0.0016 - val_acc: 0.7593 Epoch 2975/3000 1712/1712 [==============================] - 1s - loss: 7.4093e-04 - acc: 0.8271 - val_loss: 0.0020 - val_acc: 0.7290 Epoch 2976/3000 1712/1712 [==============================] - 1s - loss: 8.5300e-04 - acc: 0.8224 - val_loss: 0.0020 - val_acc: 0.7243 Epoch 2977/3000 1712/1712 [==============================] - 1s - loss: 8.3256e-04 - acc: 0.8388 - val_loss: 0.0026 - val_acc: 0.7196 Epoch 2978/3000 1712/1712 [==============================] - 1s - loss: 8.1400e-04 - acc: 0.8388 - val_loss: 0.0016 - val_acc: 0.7383 Epoch 2979/3000 1712/1712 [==============================] - 1s - loss: 8.3703e-04 - acc: 0.8400 - val_loss: 0.0014 - val_acc: 0.7500 Epoch 2980/3000 1712/1712 [==============================] - 1s - loss: 8.0414e-04 - acc: 0.8370 - val_loss: 0.0021 - val_acc: 0.7103 Epoch 2981/3000 1712/1712 [==============================] - 1s - loss: 7.8578e-04 - acc: 0.8213 - val_loss: 0.0018 - val_acc: 0.7780 Epoch 2982/3000 1712/1712 [==============================] - 1s - loss: 7.9337e-04 - acc: 0.8341 - val_loss: 0.0019 - val_acc: 0.7593 Epoch 2983/3000 1712/1712 [==============================] - 1s - loss: 8.3073e-04 - acc: 0.8306 - val_loss: 0.0022 - val_acc: 0.7500 Epoch 2984/3000 1712/1712 [==============================] - 1s - loss: 8.2631e-04 - acc: 0.8359 - val_loss: 0.0017 - val_acc: 0.7523 Epoch 2985/3000 1712/1712 [==============================] - 1s - loss: 8.0482e-04 - acc: 0.8218 - val_loss: 0.0024 - val_acc: 0.7477 Epoch 2986/3000 1712/1712 [==============================] - 1s - loss: 8.0450e-04 - acc: 0.8236 - val_loss: 0.0016 - val_acc: 0.7850 Epoch 2987/3000 1712/1712 [==============================] - 1s - loss: 8.0739e-04 - acc: 0.8411 - val_loss: 0.0012 - val_acc: 0.7967 Epoch 2988/3000 1712/1712 [==============================] - 1s - loss: 7.9976e-04 - acc: 0.8271 - val_loss: 0.0013 - val_acc: 0.7804 Epoch 2989/3000 1712/1712 [==============================] - 1s - loss: 7.5740e-04 - acc: 0.8411 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2990/3000 1712/1712 [==============================] - 1s - loss: 8.4493e-04 - acc: 0.8324 - val_loss: 0.0012 - val_acc: 0.7804 Epoch 2991/3000 1712/1712 [==============================] - 1s - loss: 7.3931e-04 - acc: 0.8440 - val_loss: 0.0014 - val_acc: 0.7710 Epoch 2992/3000 1712/1712 [==============================] - 1s - loss: 8.5842e-04 - acc: 0.8347 - val_loss: 0.0012 - val_acc: 0.8131 Epoch 2993/3000 1712/1712 [==============================] - 1s - loss: 8.2477e-04 - acc: 0.8259 - val_loss: 0.0012 - val_acc: 0.8084 Epoch 2994/3000 1712/1712 [==============================] - 1s - loss: 8.0352e-04 - acc: 0.8359 - val_loss: 0.0013 - val_acc: 0.8037 Epoch 2995/3000 1712/1712 [==============================] - 1s - loss: 8.1648e-04 - acc: 0.8405 - val_loss: 0.0014 - val_acc: 0.7477 Epoch 2996/3000 1712/1712 [==============================] - 1s - loss: 8.1849e-04 - acc: 0.8324 - val_loss: 0.0012 - val_acc: 0.8014 Epoch 2997/3000 1712/1712 [==============================] - 1s - loss: 7.5974e-04 - acc: 0.8289 - val_loss: 0.0013 - val_acc: 0.7850 Epoch 2998/3000 1712/1712 [==============================] - 1s - loss: 7.8131e-04 - acc: 0.8388 - val_loss: 0.0013 - val_acc: 0.8201 Epoch 2999/3000 1712/1712 [==============================] - 1s - loss: 8.1751e-04 - acc: 0.8376 - val_loss: 0.0014 - val_acc: 0.7827 Epoch 3000/3000 1712/1712 [==============================] - 1s - loss: 7.4825e-04 - acc: 0.8429 - val_loss: 0.0013 - val_acc: 0.7874
(IMPLEMENTATION) Answer a few questions and visualize the loss¶
Question 1: Outline the steps you took to get to your final neural network architecture and your reasoning at each step.
Answer:
- I added 3 convolution layers. I decrease kernel size on every layer but increase dimentionality, so model could generalize data better.
- After each convolution layer I added Max pooling layer. I remember that it's not recommended to add max pooling after first convolution layer but image size is big, so I'm using max pooling to manage model size.
- After convolution layers I flattened my data and added 2 fully connected layers. I tried different activation functions. It looks like 'tanh' works best with signed data. Final fully connected layer should have dimentionality 30 to provide output points.
- I added Dropout layers to control overfitting.
Question 2: Defend your choice of optimizer. Which optimizers did you test, and how did you determine which worked best?
Answer: I tried multiple optimizers: Adam optimizer, NAdam optimizer, Stochastic Gradient Descent and RMSProp. I tested optimizers with different parameters on 100 epochs. Surprisingly, RMSProp with default settings performs better than other optimizers (not much better but still better). So I decided to use RMSProp to train on 3000 epochs. 3000 epochs looks like overkill because I don't see significant improvement after 1000 epochs.
Use the code cell below to plot the training and validation loss of your neural network. You may find this resource useful.
In [61]:
## Visualize the training and validation loss of your neural network
plt.plot(history.history['acc'])
plt.plot(history.history['val_acc'])
plt.title('model accuracy')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['train', 'test'], loc='upper left')
plt.show()
# summarize history for loss
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.title('model loss')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['train', 'test'], loc='upper left')
plt.show()
Question 3: Do you notice any evidence of overfitting or underfitting in the above plot? If so, what steps have you taken to improve your model? Note that slight overfitting or underfitting will not hurt your chances of a successful submission, as long as you have attempted some solutions towards improving your model (such as regularization, dropout, increased/decreased number of layers, etc).
Answer: I see some signs of overfitting. Model performs better on train data than on test data. Aslo mode performs significantly better on some buckets. Maybe I can increase dropout size or size of buckets to improve this.
Visualize a Subset of the Test Predictions¶
Execute the code cell below to visualize your model's predicted keypoints on a subset of the testing images.
In [62]:
y_test = model.predict(X_test)
fig = plt.figure(figsize=(20,20))
fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)
for i in range(9):
ax = fig.add_subplot(3, 3, i + 1, xticks=[], yticks=[])
plot_data(X_test[i], y_test[i], ax)
Step 8: Complete the pipeline¶
With the work you did in Sections 1 and 2 of this notebook, along with your freshly trained facial keypoint detector, you can now complete the full pipeline. That is given a color image containing a person or persons you can now
- Detect the faces in this image automatically using OpenCV
- Predict the facial keypoints in each face detected in the image
- Paint predicted keypoints on each face detected
In this Subsection you will do just this!
(IMPLEMENTATION) Facial Keypoints Detector¶
Use the OpenCV face detection functionality you built in previous Sections to expand the functionality of your keypoints detector to color images with arbitrary size. Your function should perform the following steps
- Accept a color image.
- Convert the image to grayscale.
- Detect and crop the face contained in the image.
- Locate the facial keypoints in the cropped image.
- Overlay the facial keypoints in the original (color, uncropped) image.
Note: step 4 can be the trickiest because remember your convolutional network is only trained to detect facial keypoints in $96 \times 96$ grayscale images where each pixel was normalized to lie in the interval $[0,1]$, and remember that each facial keypoint was normalized during training to the interval $[-1,1]$. This means - practically speaking - to paint detected keypoints onto a test face you need to perform this same pre-processing to your candidate face - that is after detecting it you should resize it to $96 \times 96$ and normalize its values before feeding it into your facial keypoint detector. To be shown correctly on the original image the output keypoints from your detector then need to be shifted and re-normalized from the interval $[-1,1]$ to the width and height of your detected face.
When complete you should be able to produce example images like the one below
In [63]:
# Load in color image for face detection
image = cv2.imread('images/obamas4.jpg')
# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
image_copy = np.copy(image)
# plot our image
fig = plt.figure(figsize = (9,9))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('image copy')
ax1.imshow(image_copy)
Out[63]:
<matplotlib.image.AxesImage at 0x7fcfc269d898>
In [81]:
### TODO: Use the face detection code we saw in Section 1 with your trained conv-net
## TODO : Paint the predicted keypoints on the test image
# Convert the RGB image to grayscale
gray = cv2.cvtColor(image_copy, cv2.COLOR_RGB2GRAY)
# Extract the pre-trained face detector from an xml file
face_cascade = cv2.CascadeClassifier('detector_architectures/haarcascade_frontalface_default.xml')
# Detect the faces in image
faces = face_cascade.detectMultiScale(gray, 1.25, 6)
# Print the number of faces detected in the image
print('Number of faces detected:', len(faces))
# Make a copy of the orginal image to draw face detections on
image_with_detections = np.copy(image)
# Get the bounding box for each detected face
for (x,y,w,h) in faces:
# Add a red bounding box to the detections image
cv2.rectangle(image_with_detections, (x,y), (x+w,y+h), (255,0,0), 3)
for (x,y,w,h) in faces:
# Get face and normalize value
face_gray = gray[y:y+h, x:x+w] / 255
# Resize to 96x96 and add 2 dimensions, so final input size will be (1, 96, 96, 1)
resized_face = np.expand_dims(np.resize(face_gray,(96,96)), axis=2)
points = model.predict(np.expand_dims(resized_face, axis=0))
# Render 15 points
for i in range(15):
x_point = points[0, i * 2]
y_point = points[0, i * 2 + 1]
cv2.circle(image_with_detections, (x + int(x_point * h/2 + h/2), y + int(y_point * w/2 + w/2)), 3, (0, 255, 0) , -1)
# Display the image with the detections
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Image with Face Detections and keypoints')
ax1.imshow(image_with_detections)
Number of faces detected: 2
Out[81]:
<matplotlib.image.AxesImage at 0x7fcfd4078e48>
(Optional) Further Directions - add a filter using facial keypoints to your laptop camera¶
Now you can add facial keypoint detection to your laptop camera - as illustrated in the gif below.
The next Python cell contains the basic laptop video camera function used in the previous optional video exercises. Combine it with the functionality you developed for keypoint detection and marking in the previous exercise and you should be good to go!
In [ ]:
import cv2
import time
from keras.models import load_model
def laptop_camera_go():
# Create instance of video capturer
cv2.namedWindow("face detection activated")
vc = cv2.VideoCapture(0)
# Try to get the first frame
if vc.isOpened():
rval, frame = vc.read()
else:
rval = False
# keep video stream open
while rval:
# plot image from camera with detections marked
cv2.imshow("face detection activated", frame)
# exit functionality - press any key to exit laptop video
key = cv2.waitKey(20)
if key > 0: # exit by pressing any key
# destroy windows
cv2.destroyAllWindows()
# hack from stack overflow for making sure window closes on osx --> https://stackoverflow.com/questions/6116564/destroywindow-does-not-close-window-on-mac-using-python-and-opencv
for i in range (1,5):
cv2.waitKey(1)
return
# read next frame
time.sleep(0.05) # control framerate for computation - default 20 frames per sec
rval, frame = vc.read()
In [ ]:
# Run your keypoint face painter
laptop_camera_go()
(Optional) Further Directions - add a filter using facial keypoints¶
Using your freshly minted facial keypoint detector pipeline you can now do things like add fun filters to a person's face automatically. In this optional exercise you can play around with adding sunglasses automatically to each individual's face in an image as shown in a demonstration image below.
To produce this effect an image of a pair of sunglasses shown in the Python cell below.
In [ ]:
# Load in sunglasses image - note the usage of the special option
# cv2.IMREAD_UNCHANGED, this option is used because the sunglasses
# image has a 4th channel that allows us to control how transparent each pixel in the image is
sunglasses = cv2.imread("images/sunglasses_4.png", cv2.IMREAD_UNCHANGED)
# Plot the image
fig = plt.figure(figsize = (6,6))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.imshow(sunglasses)
ax1.axis('off');
This image is placed over each individual's face using the detected eye points to determine the location of the sunglasses, and eyebrow points to determine the size that the sunglasses should be for each person (one could also use the nose point to determine this).
Notice that this image actually has 4 channels, not just 3.
In [ ]:
# Print out the shape of the sunglasses image
print ('The sunglasses image has shape: ' + str(np.shape(sunglasses)))
It has the usual red, blue, and green channels any color image has, with the 4th channel representing the transparency level of each pixel in the image. Here's how the transparency channel works: the lower the value, the more transparent the pixel will become. The lower bound (completely transparent) is zero here, so any pixels set to 0 will not be seen.
This is how we can place this image of sunglasses on someone's face and still see the area around of their face where the sunglasses lie - because these pixels in the sunglasses image have been made completely transparent.
Lets check out the alpha channel of our sunglasses image in the next Python cell. Note because many of the pixels near the boundary are transparent we'll need to explicitly print out non-zero values if we want to see them.
In [ ]:
# Print out the sunglasses transparency (alpha) channel
alpha_channel = sunglasses[:,:,3]
print ('the alpha channel here looks like')
print (alpha_channel)
# Just to double check that there are indeed non-zero values
# Let's find and print out every value greater than zero
values = np.where(alpha_channel != 0)
print ('\n the non-zero values of the alpha channel look like')
print (values)
This means that when we place this sunglasses image on top of another image, we can use the transparency channel as a filter to tell us which pixels to overlay on a new image (only the non-transparent ones with values greater than zero).
One last thing: it's helpful to understand which keypoint belongs to the eyes, mouth, etc. So, in the image below, we also display the index of each facial keypoint directly on the image so that you can tell which keypoints are for the eyes, eyebrows, etc.
With this information, you're well on your way to completing this filtering task! See if you can place the sunglasses automatically on the individuals in the image loaded in / shown in the next Python cell.
In [ ]:
# Load in color image for face detection
image = cv2.imread('images/obamas4.jpg')
# Convert the image to RGB colorspace
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
# Plot the image
fig = plt.figure(figsize = (8,8))
ax1 = fig.add_subplot(111)
ax1.set_xticks([])
ax1.set_yticks([])
ax1.set_title('Original Image')
ax1.imshow(image)
In [ ]:
## (Optional) TODO: Use the face detection code we saw in Section 1 with your trained conv-net to put
## sunglasses on the individuals in our test image
(Optional) Further Directions - add a filter using facial keypoints to your laptop camera¶
Now you can add the sunglasses filter to your laptop camera - as illustrated in the gif below.
The next Python cell contains the basic laptop video camera function used in the previous optional video exercises. Combine it with the functionality you developed for adding sunglasses to someone's face in the previous optional exercise and you should be good to go!
In [ ]:
import cv2
import time
from keras.models import load_model
import numpy as np
def laptop_camera_go():
# Create instance of video capturer
cv2.namedWindow("face detection activated")
vc = cv2.VideoCapture(0)
# try to get the first frame
if vc.isOpened():
rval, frame = vc.read()
else:
rval = False
# Keep video stream open
while rval:
# Plot image from camera with detections marked
cv2.imshow("face detection activated", frame)
# Exit functionality - press any key to exit laptop video
key = cv2.waitKey(20)
if key > 0: # exit by pressing any key
# Destroy windows
cv2.destroyAllWindows()
for i in range (1,5):
cv2.waitKey(1)
return
# Read next frame
time.sleep(0.05) # control framerate for computation - default 20 frames per sec
rval, frame = vc.read()
In [ ]:
# Load facial landmark detector model
model = load_model('my_model.h5')
# Run sunglasses painter
laptop_camera_go()